Skip to content
TheChatArchiveThe Chat Archive
gpt-4o-2024-05-13· OpenAIAi2 WildChat-4.8M (AI2)

write for master thesis, correct if neccesary: 1. Multivariate analysis 1.1.…

  1. User

    write for master thesis, correct if neccesary: 1. Multivariate analysis

    1.1. Introduction

    In today’s increasingly complex and competitive business environment, understanding the multifaceted relationships between various financial and non-financial metrics is crucial for informed decision-making. Multivariate analysis offers a robust framework for examining such intricate dynamics, allowing researchers and professionals to simultaneously consider multiple variables and their interdependencies. This analytical approach not only provides deeper insights into the underlying patterns but also enhances the predictive accuracy of the models. The objective of this study is to employ multivariate analysis to explore and quantify the determinants impacting a critical financial measure, PD2023 (presumably a performance metric for the year 2023). By analyzing a comprehensive dataset that includes both traditional financial metrics and environmental performance indicators, this study aims to identify key determinants and understand their relationships with PD2023. Specifically, the analysis will focus on the following:

    1. Control variables: Previous year’s performance (PD2022), firm size (size_2023), leverage (LV_2023), return on assets (ROA_2023), capital expenditure (CAPEX_2023), and other control variables that are traditionally considered influential in financial performance analysis.
    2. physical risk (composite score at four scenarios: low, medium, highmedium and high levels of physical risk)in 3 time horizons 2030, 2040 and 2050 … write this part
    3. Environmental Performance: Absolute emissions (LN_EMISIONES_2022) and relative emissions (LN_EMISSIONSREL_2022). These variables are particularly important in the context of increasing regulatory scrutiny and stakeholder interest in corporate sustainability practices.
    4. Enviromental initiative (ENV_INV_INI_2022) and compensation for sustainabitlity (COMP_SUST_2022) variables … write this part 1.1.1. Hypotheses and Analytical Approach The hypotheses driving this study are centered around the anticipated impact of financial metrics and environmental performance on PD2023:
    • H1: Control variables such as PD2022, size_2023, LV_2023, ROA_2023, and CAPEX_2023 significantly influence PD2023.
    • H2: if physical risk impact in PD write this part
    • H3: emission and environmental actions. Seen if the effect of physical risk is different for companies with more active in terms of environmental responsibility To test these hypotheses, this study estimates a series of twelve multivariate regression models. The models vary in complexity, with some focusing solely on financial metrics and others incorporating emission variables. This approach enables the identification of consistent patterns and robust relationships, facilitating a nuanced understanding of how different factors contribute to financial performance. 1.2. Estimated models 1.2.1. General Model with Control variables: Model 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + Asset_turnover_2023 + CAPEX_2023 Model 2: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + CAPEX_2023 Model 3: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 Model 4: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023

    Results: Model Model 1 Model 2 Model 3 Model 4 intercept 0.0783671*** 0.0621015** 0.0694160*** PD2022 0.5111060*** 0.5106081*** 0.5128044*** 0.5097252*** size_2023 -0.0056918*** -0.0050308*** -0.0055663*** -0.0009534** BTM_2023 0.0043432*** 0.0043876*** 0.0044810*** 0.0043743*** LV_2023 0.1292126*** 0.1293215*** 0.1371421*** 0.1428434*** ROA_2023 -0.0901572*** -0.0954977*** -0.0875209*** -0.0913884*** Asset_turnover_2023 -0.0095013* CAPEX_2023 0.1267542* 0.1320588* R-squared 0.4114 0.4104 0.4088 0.4733 ***, **, *, . denote significance at the 0.1%, 1%, 5% and 10%, respectively.

    Interpretation:

    1. Intercept: The intercept shows statistical significance at the 10% level in Models 2 and 9, and at the 5% level in Model 3. In the other models, it is not significant. This suggests that the base level of the dependent variable (when all predictors are zero) varies and that additional factors might need consideration.
    2. PD2022: The coefficient for PD2022 is highly consistent and statistically significant at the 1% level across all models, with values around 0.49-0.50. This indicates a strong and highly stable positive relationship between PD2022 and the dependent variable, suggesting that PD2022 is a crucial determinant in this analysis.
    3. size_2023: The coefficient for size_2023 is negative and statistically significant across several models, particularly strong with effects in Models 1, 2, 3, 4, 8, and 9 at varying levels of significance (1% - 5%). This implies that size in 2023 has a negative impact on the dependent variable.
    4. BTM_2023: The BTM_2023 coefficient is not significant in most models, indicating that this variable might have a negligible impact on the dependent variable.
    5. LV_2023: The coefficient for LV_2023 (leverage) is positive and highly significant across all models (1% level). This suggests a robust and positive relationship between leverage and the dependent variable.
    6. ROA_2023: The ROA_2023 coefficient is negative and statistically significant at the 1% level across all models. This suggests a strong negative relationship between return on assets and the dependent variable.
    7. Asset_turnover_2023: Asset_turnover_2023 appears in some models with positive coefficients but it’s not statistically significant, indicating limited or no impact on the dependent variable.
    8. CAPEX_2023: CAPEX_2023 shows positive and statistically significant coefficients in several models (1%, 5%, and 10% levels). The relationship between capital expenditure and the dependent variable is strongly positive, highlighting its importance.
    9. R-squared: The R-squared values range from 0.3864 to 0.4499, indicating that the models explain roughly 38.64% to 44.99% of the variability in the dependent variable. Models 4, 8, and 12 have higher R-squared values, suggesting a better fit compared to others. Correct Conclusion: Model 3 and model 4 is better, so we will use this model for continues analysis 1.2.2. Model including pyshical risk scenarios Model 3 Scenario 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30 Model 3 Scenario 2: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_30 Model 3 Scenario 3: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_30 Model 3 Scenario 4: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_30 Model 3 Scenario 5: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_40 Model 3 Scenario 6: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_40 Model 3 Scenario 7: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_40 Model 3 Scenario 8: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_40 Model 3 Scenario 9: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_50 Model 3 Scenario 10: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_50 Model 3 Scenario 11: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_50 Model 3 Scenario 12: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_50 Model 4 Scenario 1: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30 Model 4 Scenario 2: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_30 Model 4 Scenario 3: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_30 Model 4 Scenario 4: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_30 Model 4 Scenario 5: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_40 Model 4 Scenario 6: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_40 Model 4 Scenario 7: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_40 Model 4 Scenario 8: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_40 Model 4 Scenario 9: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_50 Model 4 Scenario 10: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUM_50 Model 4 Scenario 11: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + MEDIUMHIGH_50 Model 4 Scenario 12: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + HIGH_50 Result: Model Model 3. Score 1 Model 3. Score 2 Model 3. Score 3 Model 3. Score 4 Model 3. Score 5 Model 3. Score 6 Model 3. Score 7 Model 3. Score 8 Model 3. Score 9 Model 3. Score 10 Model 3. Score 11 Model 3. Score 12 intercept 0.1333379*** 0.1312233*** 0.1359114*** 0.1338333*** 0.1390164*** 0.1372199*** 0.1405299*** 0.1433760*** 0.1399665*** 0.1421692*** 0.1519492*** 0.1537219*** PD2022 0.5154807*** 0.5155672*** 0.5154683*** 0.5156095*** 0.5150503*** 0.5152366*** 0.5149512*** 0.5148515*** 0.5151348*** 0.5148965*** 0.5148420*** 0.5147547*** size_2023 -0.0053289*** -0.0053181*** -0.0053093*** -0.0053141*** -0.0053134*** -0.0053146*** -0.0053057*** -0.0052735*** -0.0053068*** -0.0052768*** -0.0052854*** -0.0052702*** BTM_2023 0.0045383*** 0.0045367*** 0.0045431*** 0.0045455*** 0.0045460*** 0.0045421*** 0.0045462*** 0.0045477*** 0.0045454*** 0.0045466*** 0.0045475*** 0.0045481*** LV_2023 0.1377720*** 0.1378585*** 0.1377943*** 0.1378692*** 0.1380262*** 0.1379483*** 0.1379325*** 0.1379518*** 0.1379379*** 0.1379927*** 0.1379244*** 0.1379152*** ROA_2023 -0.0847510*** -0.0845869*** -0.0846011*** -0.0844892*** -0.0845821*** -0.0846266*** -0.0847131*** -0.0847733*** -0.0846537*** -0.0847370*** -0.0847662*** -0.0849210*** LOW_30 -0.0013004*** MEDIUM_30 -0.0012842*** MEDIUMHIGH_30 -0.0013425*** HIGH_30 -0.0012962*** LOW_40 -0.0013545*** MEDIUM_40 -0.0013134*** MEDIUMHIGH_40 -0.0013347*** HIGH_40 -0.0013539*** LOW_50 -0.0013526*** MEDIUM_50 -0.0013437*** MEDIUMHIGH_50 -0.0014363*** HIGH_50 -0.0014242*** R-squared 0.4124 0.4125 0.4127 0.4126 0.4127 0.4126 0.4125 0.4126 0.4127 0.4125 0.4127 0.4126

    Interpretation: How affect the different scenarios to PD… write this part How affect intercept… write this part Conclusion: Write this part

    1.2.3. Model including emissions and environmental actions Emisiones: Model 3 Scenario 1 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ LN_EMISIONES_2022 Model 3 Scenario 2 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ LN_EMISIONES_2022 Model 3 Scenario 3 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ LN_EMISIONES_2022 Model 3 Scenario 4 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ LN_EMISIONES_2022 Model 3 Scenario 5 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ LN_EMISIONES_2022 Model 3 Scenario 6 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ LN_EMISIONES_2022 Model 3 Scenario 7 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ LN_EMISIONES_2022 Model 3 Scenario 8 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ LN_EMISIONES_2022 Model 3 Scenario 9 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ LN_EMISIONES_2022 Model 3 Scenario 10 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ LN_EMISIONES_2022 Model 3 Scenario 11 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ LN_EMISIONES_2022 Model 3 Scenario 12 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ LN_EMISIONES_2022

    Model 4 Scenario 1 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ LN_EMISIONES_2022 Model 4 Scenario 2 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ LN_EMISIONES_2022 Model 4 Scenario 3 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ LN_EMISIONES_2022 Model 4 Scenario 4 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ LN_EMISIONES_2022 Model 4 Scenario 5 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ LN_EMISIONES_2022 Model 4 Scenario 6 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ LN_EMISIONES_2022 Model 4 Scenario 7 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ LN_EMISIONES_2022 Model 4 Scenario 8 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ LN_EMISIONES_2022 Model 4 Scenario 9 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ LN_EMISIONES_2022 Model 4 Scenario 10 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ LN_EMISIONES_2022 Model 4 Scenario 11 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ LN_EMISIONES_2022 Model 4 Scenario 12 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ LN_EMISIONES_2022 Emisiones relativos: Model 3 Scenario 1 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ LN_EMISSIONSREL_2022 Model 3 Scenario 2 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ LN_EMISSIONSREL_2022 Model 3 Scenario 3 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ LN_EMISSIONSREL_2022 Model 3 Scenario 4 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ LN_EMISSIONSREL_2022 Model 3 Scenario 5 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ LN_EMISSIONSREL_2022 Model 3 Scenario 6 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ LN_EMISSIONSREL_2022 Model 3 Scenario 7 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ LN_EMISSIONSREL_2022 Model 3 Scenario 8 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ LN_EMISSIONSREL_2022 Model 3 Scenario 9 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ LN_EMISSIONSREL_2022 Model 3 Scenario 10 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ LN_EMISSIONSREL_2022 Model 3 Scenario 11 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ LN_EMISSIONSREL_2022 Model 3 Scenario 12 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ LN_EMISSIONSREL_2022

    Model 4 Scenario 1 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ LN_EMISSIONSREL_2022 Model 4 Scenario 2 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ LN_EMISSIONSREL_2022 Model 4 Scenario 3 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ LN_EMISSIONSREL_2022 Model 4 Scenario 4 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ LN_EMISSIONSREL_2022 Model 4 Scenario 5 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ LN_EMISSIONSREL_2022 Model 4 Scenario 6 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ LN_EMISSIONSREL_2022 Model 4 Scenario 7 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ LN_EMISSIONSREL_2022 Model 4 Scenario 8 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ LN_EMISSIONSREL_2022 Model 4 Scenario 9 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ LN_EMISSIONSREL_2022 Model 4 Scenario 10 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ LN_EMISSIONSREL_2022 Model 4 Scenario 11 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ LN_EMISSIONSREL_2022 Model 4 Scenario 12 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ LN_EMISSIONSREL_2022 Environmental initiative: Model 3 Scenario 1 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ ENV_INV_INI_2022Y Model 3 Scenario 2 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ ENV_INV_INI_2022Y Model 3 Scenario 3 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ ENV_INV_INI_2022Y Model 3 Scenario 4 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ ENV_INV_INI_2022Y Model 3 Scenario 5 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ ENV_INV_INI_2022Y Model 3 Scenario 6 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ ENV_INV_INI_2022Y Model 3 Scenario 7 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ ENV_INV_INI_2022Y Model 3 Scenario 8 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ ENV_INV_INI_2022Y Model 3 Scenario 9 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ ENV_INV_INI_2022Y Model 3 Scenario 10 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ ENV_INV_INI_2022Y Model 3 Scenario 11 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ ENV_INV_INI_2022Y Model 3 Scenario 12 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ ENV_INV_INI_2022Y

    Model 4 Scenario 1 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 2 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 3 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 4 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 5 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 6 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 7 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 8 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 9 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 10 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 11 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Model 4 Scenario 12 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ ENV_INV_INI_2022N + ENV_INV_INI_2022Y Compensation for sustainability Model 3 Scenario 1 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ COMP_SUST_2022Y Model 3 Scenario 2 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ COMP_SUST_2022Y Model 3 Scenario 3 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ COMP_SUST_2022Y Model 3 Scenario 4 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ COMP_SUST_2022Y Model 3 Scenario 5 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ COMP_SUST_2022Y Model 3 Scenario 6 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ COMP_SUST_2022Y Model 3 Scenario 7 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ COMP_SUST_2022Y Model 3 Scenario 8 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ COMP_SUST_2022Y Model 3 Scenario 9 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ COMP_SUST_2022Y Model 3 Scenario 10 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ COMP_SUST_2022Y Model 3 Scenario 11 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ COMP_SUST_2022Y Model 3 Scenario 12 + EMISIONES_2022: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ COMP_SUST_2022Y

    Model 4 Scenario 1 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_30+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 2 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_30+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 3 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_30+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 4 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_30+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 5 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_40+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 6 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_40+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 7 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_40+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 8 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_40+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 9 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +LOW_50+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 10 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUM_50+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 11 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +MEDIUMHIGH_50+ COMP_SUST_2022N + COMP_SUST_2022Y Model 4 Scenario 12 + EMISIONES_2022: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 +HIGH_50+ COMP_SUST_2022N + COMP_SUST_2022Y Result:  

    Model Model 3. EMISIONES 1 Model 3. EMISIONES 2 Model 3. EMISIONES 3 Model 3. EMISIONES 4 Model 3. EMISIONES 5 Model 3. EMISIONES 6 Model 3. EMISIONES 7 Model 3. EMISIONES 8 Model 3. EMISIONES 9 Model 3. EMISIONES 10 Model 3. EMISIONES 11 Model 3. EMISIONES 12 intercept 0.1372033*** 0.1351360*** 0.1395774*** 0.1375316*** 0.1425669*** 0.1407266*** 0.1438981*** 0.1464936*** 0.1434529*** 0.1453627*** 0.1547703*** 0.1562412*** PD2022 0.5144815*** 0.5145766*** 0.5144936*** 0.5146283*** 0.5140971*** 0.5142724*** 0.5139911*** 0.5139113*** 0.5141750*** 0.5139472*** 0.5139057*** 0.5138230*** size_2023 -0.0046400** -0.0046385** -0.0046373** -0.0046439** -0.0046418** -0.0046447** -0.0046298** -0.0046112** -0.0046343** -0.0046095** -0.0046224** -0.0046123** BTM_2023 0.0045000*** 0.0044990*** 0.0045055*** 0.0045078*** 0.0045084*** 0.0045046*** 0.0045081*** 0.0045102*** 0.0045077*** 0.0045089*** 0.0045101*** 0.0045107*** LV_2023 0.1391847*** 0.1392490*** 0.1391730*** 0.1392403*** 0.1393932*** 0.1393148*** 0.1393128*** 0.1393065*** 0.1393113*** 0.1393550*** 0.1392808*** 0.1392627*** ROA_2023 -0.0849573*** -0.0847985*** -0.0848090*** -0.0847044*** -0.0847894*** -0.0848365*** -0.0849214*** -0.0849771*** -0.0848591*** -0.0849443*** -0.0849663*** -0.0851190*** LOW_30 -0.0012380*** MEDIUM_30 -0.0012233*** MEDIUMHIGH_30 -0.0012798*** HIGH_30 -0.0012347*** LOW_40 -0.0012920*** MEDIUM_40 -0.0012506*** MEDIUMHIGH_40 -0.0012700*** HIGH_40 -0.0012883*** LOW_50 -0.0012895*** MEDIUM_50 -0.0012781*** MEDIUMHIGH_50 -0.0013689*** HIGH_50 -0.0013544*** LN_EMISIONES_2022 -0.0017027 -0.0016810 -0.0016630 -0.0016587 -0.0016613 -0.0016580 -0.0016741 -0.0016448 -0.0016645 -0.0016567 -0.0016440 -0.0016349
    R-squared 0.4129 0.413 0.4131 0.4131 0.4132 0.413 0.4129 0.413 0.4131 0.413 0.4131 0.413

     

    Model Model 4. EMISIONES 1 Model 4. EMISIONES 2 Model 4. EMISIONES 3 Model 4. EMISIONES 4 Model 4. EMISIONES 5 Model 4. EMISIONES 6 Model 4. EMISIONES 7 Model 4. EMISIONES 8 Model 4. EMISIONES 9 Model 4. EMISIONES 10 Model 4. EMISIONES 11 Model 4. EMISIONES 12 intercept PD2022 0.4968205*** 0.4968329*** 0.4967552*** 0.4968376*** 0.4967294*** 0.4967973*** 0.4967999*** 0.4968055*** 0.4967372*** 0.4968158*** 0.4968822*** 0.49697231*** size_2023 0.0011401 0.0012372 0.0012512 0.0011852 0.0011473 0.0011092 0.0009869 0.0009765 0.0011198 0.0009504 0.0008230 0.00068561
    BTM_2023 -0.0006102 -0.0006071 -0.0006068 -0.0006066 -0.0006114 -0.0006120 -0.0006168 -0.0006174 -0.0006131 -0.0006181 -0.0006220 -0.00062495
    LV_2023 0.1982232*** 0.1984612*** 0.1984853*** 0.1983364*** 0.1982436*** 0.1981224*** 0.1978182*** 0.1977608*** 0.1981543*** 0.1977221*** 0.1974126*** 0.19707316*** ROA_2023 -0.1705435*** -0.1701236*** -0.1702031*** -0.1703078*** -0.1705158*** -0.1705844*** -0.1710610*** -0.1711579*** -0.1706126*** -0.1712265*** -0.1716460*** -0.17209274*** LOW_30 -0.0002377 MEDIUM_30 -0.0002756 MEDIUMHIGH_30 -0.0002718 HIGH_30 -0.0026845* LOW_40 -0.0002293 MEDIUM_40 -0.0002169 MEDIUMHIGH_40 -0.0001725 HIGH_40 -0.0001646 LOW_50 -0.0002169 MEDIUM_50 -0.0001584 MEDIUMHIGH_50 -0.0001142 HIGH_50 -0.00007406 LN_EMISIONES_2022 -0.0026960* -0.0026810* -0.0026849* -0.0002506 -0.0026977* -0.0026970* -0.0027104* -0.0027095* -0.0027017* -0.0027117* -0.0027248* -0.00273084*
    R-squared 0.4457 0.4458 0.4458 0.4457 0.4457 0.4457 0.4456 0.4456 0.4457 0.4456 0.4456 0.4456

    Model Model 3. EMISSIONSREL 1 Model 3. EMISSIONSREL 2 Model 3. EMISSIONSREL 3 Model 3. EMISSIONSREL 4 Model 3. EMISSIONSREL 5 Model 3. EMISSIONSREL 6 Model 3. EMISSIONSREL 7 Model 3. EMISSIONSREL 8 Model 3. EMISSIONSREL 9 Model 3. EMISSIONSREL 10 Model 3. EMISSIONSREL 11 Model 3. EMISSIONSREL 12 intercept 0.1269910*** 0.1247362*** 0.1297759*** 0.1262328*** 0.1303181*** 0.1283068*** 0.1316277*** 0.1337368*** 0.1315412*** 0.1321394*** 0.1404204*** 0.1413043*** PD2022 0.5050857*** 0.5052116*** 0.5051322*** 0.5052150*** 0.5047766*** 0.5049183*** 0.5046450*** 0.5046948*** 0.5047595*** 0.5045969*** 0.5046710*** 0.5046256*** size_2023 -0.0054980** -0.0054822** -0.0054595** -0.0054793** -0.0054586** -0.0054596** -0.0054664** -0.0054360** -0.0054476** -0.0054390** -0.0054440** -0.0054351** BTM_2023 -0.0005158 -0.0005175 -0.0005108 -0.0005081 -0.0005177 -0.0005185 -0.0005237 -0.0005244 -0.0005205 -0.0005263 -0.0005288 -0.0005320
    LV_2023 0.1870193*** 0.1870104*** 0.1869547*** 0.1869940*** 0.1870622*** 0.1869567*** 0.1868944*** 0.1866786*** 0.1869872*** 0.1868847*** 0.1867495*** 0.1865755*** ROA_2023 -0.1861718*** -0.1858455*** -0.1860314*** -0.1857991*** -0.1858847*** -0.1857293*** -0.1860043*** -0.1862306*** -0.1858442*** -0.1862215*** -0.1861325*** -0.1864395*** LOW_30 -0.0010646* MEDIUM_30 -0.0010444* MEDIUMHIGH_30 -0.0011217** HIGH_30 -0.0010431* LOW_40 -0.0010937** MEDIUM_40 -0.0010496* MEDIUMHIGH_40 -0.0010738* HIGH_40 -0.0010882* LOW_50 -0.0011020** MEDIUM_50 -0.0010685* MEDIUMHIGH_50 -0.0011522** HIGH_50 -0.0011323*
    LN_EMISSIONSREL_2022 -0.0023691 -0.0023513 -0.0023066 -0.0023305 -0.0023325 -0.0023475 -0.0023498 -0.0023128 -0.0023318 -0.0023374 -0.0023063 -0.0023053
    R-squared 0.3885 0.3886 0.3888 0.3886 0.3887 0.3885 0.3885 0.3886 0.3887 0.3885 0.3886 0.3885

      Model Model 4. EMISSIONSREL 1 Model 4. EMISSIONSREL 2 Model 4. EMISSIONSREL 3 Model 4. EMISSIONSREL 4 Model 4. EMISSIONSREL 5 Model 4. EMISSIONSREL 6 Model 4. EMISSIONSREL 7 Model 4. EMISSIONSREL 8 Model 4. EMISSIONSREL 9 Model 4. EMISSIONSREL 10 Model 4. EMISSIONSREL 11 Model 4. EMISSIONSREL 12 intercept PD2022 0.4982708*** 0.4982774*** 0.4982263*** 0.4982801*** 0.4982179*** 0.4982571*** 0.49828289*** 0.4982899*** 0.4982270*** 0.49829957*** 0.49838208*** 0.49846427*** size_2023 -0.0004997 -0.0003760 -0.0003800 -0.0004377 -0.0005123 -0.0005437 -0.00070258 -0.0007183 -0.0005497 -0.00074557 -0.00092354 -0.00107871
    BTM_2023 -0.0005483 -0.0005456 -0.0005456 -0.0005457 -0.0005494 -0.0005498 -0.00055360 -0.0005541 -0.0005506 -0.00055456 -0.00055759 -0.00055971
    LV_2023 0.1985856*** 0.1987812*** 0.1987800*** 0.1986798*** 0.1985708*** 0.1984980*** 0.19823225*** 0.1981912*** 0.1984984*** 0.19815664*** 0.19785772*** 0.19759744*** ROA_2023 -0.1766911*** -0.1762419*** -0.1763428*** -0.1764446*** -0.1767279*** -0.1767927*** -0.17732197*** -0.1773967*** -0.1768481*** -0.17747140*** -0.17799542*** -0.17843322*** LOW_30 -0.0001515 MEDIUM_30 -0.0001949 MEDIUMHIGH_30 -0.0001882 HIGH_30 -0.0001694 LOW_40 -0.0001410 MEDIUM_40 -0.0001306 MEDIUMHIGH_40 -0.00007926 HIGH_40 -0.0000724 LOW_50 -0.0001274 MEDIUM_50 -0.00006514 MEDIUMHIGH_50 -0.00001281 HIGH_50 0.00002892 LN_EMISSIONSREL_2022 -0.0026240. -0.0025818. -0.0025882. -0.0025998. -0.0026297. -0.0026369. -0.00268372. -0.0026876. -0.0026419. -0.00269508. -0.00274259. -0.00278059.
    R-squared 0.4455 0.4456 0.4456 0.4456 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455   Model Model 3. ENV_INV_INI 1 Model 3. ENV_INV_INI 2 Model 3. ENV_INV_INI 3 Model 3. ENV_INV_INI 4 Model 3. ENV_INV_INI 5 Model 3. ENV_INV_INI 6 Model 3. ENV_INV_INI 7 Model 3. ENV_INV_INI 8 Model 3. ENV_INV_INI 9 Model 3. ENV_INV_INI 10 Model 3. ENV_INV_INI 11 Model 3. ENV_INV_INI 12 intercept 0.09181814** 0.08955373** 0.09427877** 0.09187821** 0.09580378** 0.09398086** 0.09765361** 0.09934115** 0.09688006** 0.09834841** 0.10747123** 0.1083332** PD2022 0.52620556*** 0.52624538*** 0.52616975*** 0.52626514*** 0.52576702*** 0.52594113*** 0.52566443*** 0.52566269*** 0.52578115*** 0.52560058*** 0.52559636*** 0.5255627*** size_2023 -0.00321405* -0.00320779* -0.00318586. -0.00320487. -0.00317520. -0.00317908. -0.00316930. -0.00313889. -0.00316592. -0.00313878. -0.00313815. -0.0031261.
    BTM_2023 -0.00004537 -0.00004609 -0.00003865 -0.00003416 -0.00004748 -0.00004763 -0.00005396 -0.00005278 -0.00005062 -0.00005494 -0.00005784 -0.0000606
    LV_2023 0.17271723*** 0.17280254*** 0.17279489*** 0.17284058*** 0.17286369*** 0.17278426*** 0.17276424*** 0.17266728*** 0.17276698*** 0.17277580*** 0.17268178*** 0.1725444*** ROA_2023 -0.09077967*** -0.09062646*** -0.09064974*** -0.09052490*** -0.09069279*** -0.09067223*** -0.09071497*** -0.09079697*** -0.09067613*** -0.09084464*** -0.09082943*** -0.0908865*** LOW_30 -0.00118256** MEDIUM_30 -0.00116086** MEDIUMHIGH_30 -0.00122735** HIGH_30 -0.00117155** LOW_40 -0.00121719** MEDIUM_40 -0.00117556** MEDIUMHIGH_40 -0.00120688** HIGH_40 -0.00120919** LOW_50 -0.00122014** MEDIUM_50 -0.00120218** MEDIUMHIGH_50 -0.00129351** HIGH_50 -0.0012697** ENV_INV_INI_2022Y -0.01079555 -0.01073313 -0.01068283 -0.01063678 -0.01076058 -0.01080412 -0.01089093 -0.01086163 -0.01080057 -0.01090849 -0.01085645 -0.0109473
    R-squared 0.3839 0.3839 0.3841 0.384 0.3841 0.3839 0.3839 0.3839 0.384 0.3839 0.3841 0.3839

      Model Model 4. ENV_INV_INI 1 Model 4. ENV_INV_INI 2 Model 4. ENV_INV_INI 3 Model 4. ENV_INV_INI 4 Model 4. ENV_INV_INI 5 Model 4. ENV_INV_INI 6 Model 4. ENV_INV_INI 7 Model 4. ENV_INV_INI 8 Model 4. ENV_INV_INI 9 Model 4. ENV_INV_INI 10 Model 4. ENV_INV_INI 11 Model 4. ENV_INV_INI 12 intercept PD2022 0.52620556*** 0.52624538*** 0.52616975*** 0.52626514*** 0.52576702*** 0.52594113*** 0.52566443*** 0.52566269*** 0.52578115*** 0.52560058*** 0.52559636*** 0.5255627*** size_2023 -0.00321405* -0.00320779* -0.00318586. -0.00320487. -0.00317520. -0.00317908. -0.00316930. -0.00313889. -0.00316592. -0.00313878. -0.00313815. -0.0031261.
    BTM_2023 -0.00004537 -0.00004609 -0.00003865 -0.00003416 -0.00004748 -0.00004763 -0.00005396 -0.00005278 -0.00005062 -0.00005494 -0.00005784 -0.0000606
    LV_2023 0.17271723*** 0.17280254*** 0.17279489*** 0.17284058*** 0.17286369*** 0.17278426*** 0.17276424*** 0.17266728*** 0.17276698*** 0.17277580*** 0.17268178*** 0.1725444*** ROA_2023 -0.09077967*** -0.09062646*** -0.09064974*** -0.09052490*** -0.09069279*** -0.09067223*** -0.09071497*** -0.09079697*** -0.09067613*** -0.09084464*** -0.09082943*** -0.0908865*** LOW_30 -0.00118256** MEDIUM_30 -0.00116086** MEDIUMHIGH_30 -0.00122735** HIGH_30 -0.00117155** LOW_40 -0.00121719** MEDIUM_40 -0.00117556** MEDIUMHIGH_40 -0.00120688** HIGH_40 -0.00120919** LOW_50 -0.00122014** MEDIUM_50 -0.00120218** MEDIUMHIGH_50 -0.00129351** HIGH_50 -0.0012697** ENV_INV_INI_2022N 0.09181814** 0.08955373** 0.09427877** 0.09187821** 0.09580378** 0.09398086** 0.09765361** 0.09934115** 0.09688006** 0.09834841** 0.10747123** 0.1083332** ENV_INV_INI_2022Y 0.08102260* 0.07882060* 0.08359594* 0.08124143* 0.08504320* 0.08317673* 0.08676268* 0.08847952** 0.08607949* 0.08743992* 0.09661479** 0.0973859** R-squared 0.4465 0.4466 0.4467 0.4466 0.4467 0.4466 0.4465 0.4466 0.4467 0.4465 0.4467 0.4466

    Model Model 3. COMP_SUST 1 Model 3. COMP_SUST 2 Model 3. COMP_SUST 3 Model 3. COMP_SUST 4 Model 3. COMP_SUST 5 Model 3. COMP_SUST 6 Model 3. COMP_SUST 7 Model 3. COMP_SUST 8 Model 3. COMP_SUST 9 Model 3. COMP_SUST 10 Model 3. COMP_SUST 11 Model 3. COMP_SUST 12 intercept 0.1007621*** 0.09835028*** 0.10313436*** 0.10069735*** 0.10480646*** 0.1029617*** 0.10683185*** 0.10853926*** 0.10594653*** 0.10756161*** 0.11681148*** 0.1178206*** PD2022 0.5278887*** 0.52791760*** 0.52783041*** 0.52792332*** 0.52742695*** 0.5276140*** 0.52734313*** 0.52733384*** 0.52745122*** 0.52727732*** 0.52726256*** 0.5272406*** size_2023 -0.0040762* -0.00406212* -0.00403602* -0.00405027* -0.00403326* -0.0040398* -0.00404029* -0.00400345* -0.00403036* -0.00400810* -0.00400169* -0.0039957*
    BTM_2023 0.0000838 0.00008199 0.00008903 0.00009323 0.00008093 0.0000813 0.00007604 0.00007652 0.00007862 0.00007489 0.00007097 0.0000690
    LV_2023 0.1723753*** 0.17246645*** 0.17246024*** 0.17251504*** 0.17251936*** 0.1724407*** 0.17240873*** 0.17230889*** 0.17242089*** 0.17241537*** 0.17232514*** 0.1721716*** ROA_2023 -0.0912498*** -0.09108856*** -0.09111133*** -0.09097113*** -0.09117036*** -0.0911450*** -0.09120279*** -0.09128829*** -0.09115008*** -0.09134472*** -0.09132262*** -0.0913965*** LOW_30 -0.0012247** MEDIUM_30 -0.00120191** MEDIUMHIGH_30 -0.00126914** HIGH_30 -0.00121329** LOW_40 -0.00125914** MEDIUM_40 -0.0012165** MEDIUMHIGH_40 -0.00124870** HIGH_40 -0.00125093** LOW_50 -0.00126199** MEDIUM_50 -0.00124386** MEDIUMHIGH_50 -0.00133580** HIGH_50 -0.0013121** COMP_SUST_2022Y 0.0050404 0.00496997 0.00494940 0.00492037 0.00500833 0.0050187 0.00510544 0.00504067 0.00508592 0.00507363 0.00501398 0.0050355
    R-squared 0.3833 0.3834 0.3836 0.3835 0.3835 0.3834 0.3834 0.3834 0.3835 0.3833 0.3835 0.3834   Model Model 4. COMP_SUST 1 Model 4. COMP_SUST 2 Model 4. COMP_SUST 3 Model 4. COMP_SUST 4 Model 4. COMP_SUST 5 Model 4. COMP_SUST 6 Model 4. COMP_SUST 7 Model 4. COMP_SUST 8 Model 4. COMP_SUST 9 Model 4. COMP_SUST 10 Model 4. COMP_SUST 11 Model 4. COMP_SUST 12 intercept PD2022 0.5278887*** 0.52791760*** 0.52783041*** 0.52792332*** 0.52742695*** 0.5276140*** 0.52734313*** 0.52733384*** 0.52745122*** 0.52727732*** 0.52726256*** 0.5272406*** size_2023 -0.0040762* -0.00406212* -0.00403602* -0.00405027* -0.00403326* -0.0040398* -0.00404029* -0.00400345* -0.00403036* -0.00400810* -0.00400169* -0.0039957*
    BTM_2023 0.0000838 0.00008199 0.00008903 0.00009323 0.00008093 0.0000813 0.00007604 0.00007652 0.00007862 0.00007489 0.00007097 0.0000690
    LV_2023 0.1723753*** 0.17246645*** 0.17246024*** 0.17251504*** 0.17251936*** 0.1724407*** 0.17240873*** 0.17230889*** 0.17242089*** 0.17241537*** 0.17232514*** 0.1721716*** ROA_2023 -0.0912498*** -0.09108856*** -0.09111133*** -0.09097113*** -0.09117036*** -0.0911450*** -0.09120279*** -0.09128829*** -0.09115008*** -0.09134472*** -0.09132262*** -0.0913965*** LOW_30 -0.0012247** MEDIUM_30 -0.00120191** MEDIUMHIGH_30 -0.00126914** HIGH_30 -0.00121329** LOW_40 -0.00125914** MEDIUM_40 -0.0012165** MEDIUMHIGH_40 -0.00124870** HIGH_40 -0.00125093** LOW_50 -0.00126199** MEDIUM_50 -0.00124386** MEDIUMHIGH_50 -0.00133580** HIGH_50 -0.0013121** COMP_SUST_2022N 0.1007621*** 0.09835028*** 0.10313436*** 0.10069735*** 0.10480646*** 0.1029617*** 0.10683185*** 0.10853926*** 0.10594653*** 0.10756161*** 0.11681148*** 0.1178206*** COMP_SUST_2022Y 0.1058025*** 0.10332025*** 0.10808377*** 0.10561772*** 0.10981479*** 0.1079804*** 0.11193729*** 0.11357994*** 0.11103245*** 0.11263524*** 0.12182546*** 0.1228561*** R-squared 0.446 0.446 0.4462 0.4461 0.4461 0.446 0.446 0.446 0.4461 0.446 0.4462 0.446

    Interpretation: emission and environmental actions. Seen if the effect of physical risk is different for companies with more active in terms of environmental responsibility … write this part How affect intercept… write this part Conclusion: Write this part 2. Conclusion

    This analysis provides a comprehensive examination of the determinants impacting PD2023, with a specific focus on emissions variables. By estimating 12 different models, we have identified several key insights and robust findings. 2.1 PD2022 and control variables

    1. PD2022: Consistently positive and highly significant across all models, indicating a strong predictive power for PD2023.
    2. size_2023: Generally exhibits a negative and significant relationship with PD2023, suggesting larger firm size tends to reduce PD2023.
    3. LV_2023 (Leverage): Shows a positive and significant impact across all models, highlighting that higher leverage is associated with higher PD2023.
    4. ROA_2023 (Return on Assets): Demonstrates a negative and significant relationship, implying that higher profitability is linked with lower PD2023.
    5. CAPEX_2023 (Capital Expenditure): Positive and significant in many models, indicating that higher capital expenditure is positively correlated with PD2023. 2.2. Physical escecearios Write this part 2.2 PD2022 Emissions and environmental actions Variables:

    Absolute Emissions (LN_EMISIONES_2022):

    • Models 5-8 incorporate LN_EMISIONES_2022.
    • Consistently negative and significant coefficients across these models indicate that higher absolute emissions in 2022 are associated with lower PD2023.
    • Significance levels increase from 10% to 1% as models are streamlined, reinforcing the robustness of this relationship.

    Relative Emissions (LN_EMISSIONSREL_2022):

    • Models 9-12 incorporate LN_EMISSIONSREL_2022.
    • Similar to absolute emissions, LN_EMISSIONSREL_2022 shows a consistent negative and significant relationship with PD2023.
    • Increasing significance from 10% to 1% across models indicates that higher relative emissions also detract from PD2023.

    Write this part 2.3 Overall Insights

    1. Strong Predictors: Key determinants such as PD2022, LV_2023, ROA_2023, and CAPEX_2023 consistently show significant impacts on PD2023, highlighting their importance in predictive models.
    2. Emissions Impact: The negative relationships with PD2023 for both absolute emissions (LN_EMISIONES_2022) and relative emissions (LN_EMISSIONSREL_2022) underscore the importance of environmental performance on financial outcomes.
    3. Model Robustness: The stability and significance of emissions variables across different model configurations imply that emissions are an indispensable component of understanding PD2023 dynamics.

    2.4 Concluding Remarks This study underscores the multi-faceted nature of determinants affecting PD2023, emphasizing both traditional financial metrics and environmental performance. The consistent negative impact of emissions highlights the growing imperative for sustainable business practices. Companies should prioritize reducing emissions and improving environmental performance to potentially achieve better financial outcomes and enhance stakeholder trust. These insights can guide strategic decisions, policy-making, and further research, advancing the understanding of the complex interplay between environmental factors and financial performance.

  2. Assistant

    Certainly! Here’s a revised and expanded version of your master thesis draft for the sections that needed additions and corrections:


    1. Multivariate Analysis

    1.1. Introduction

    In today’s increasingly complex and competitive business environment, understanding the multifaceted relationships between various financial and non-financial metrics is crucial for informed decision-making. Multivariate analysis offers a robust framework for examining such intricate dynamics, allowing researchers and professionals to simultaneously consider multiple variables and their interdependencies. This analytical approach not only provides deeper insights into the underlying patterns but also enhances the predictive accuracy of the models.

    The objective of this study is to employ multivariate analysis to explore and quantify the determinants impacting a critical financial measure, PD2023 (presumably a performance metric for the year 2023). By analyzing a comprehensive dataset that includes both traditional financial metrics and environmental performance indicators, this study aims to identify key determinants and understand their relationships with PD2023. Specifically, the analysis will focus on the following:

    1. Control Variables: Previous year’s performance (PD2022), firm size (size_2023), leverage (LV_2023), return on assets (ROA_2023), capital expenditure (CAPEX_2023), and other control variables that are traditionally considered influential in financial performance analysis.
    2. Physical Risk: Composite scores at four scenarios (low, medium, medium-high, and high levels of physical risk) across three time horizons: 2030, 2040, and 2050. This section will delve into how varying levels of physical risk at different temporal stages might affect PD2023.
    3. Environmental Performance: Absolute emissions (LN_EMISIONES_2022) and relative emissions (LN_EMISSIONSREL_2022). These variables are particularly important in the context of increasing regulatory scrutiny and stakeholder interest in corporate sustainability practices.
    4. Environmental Initiatives and Compensation for Sustainability: These variables (ENV_INV_INI_2022 and COMP_SUST_2022) will capture the impact of companies' proactive environmental measures and sustainability-related compensation practices on PD2023.

    1.1.1. Hypotheses and Analytical Approach

    The hypotheses driving this study are centered around the anticipated impact of financial metrics and environmental performance on PD2023:

    • H1: Control variables such as PD2022, size_2023, LV_2023, ROA_2023, and CAPEX_2023 significantly influence PD2023.
    • H2: Physical risk levels (low, medium, medium-high, high) in different time horizons (2030, 2040, 2050) have a significant impact on PD2023.
    • H3: Emissions and environmental actions significantly influence PD2023, with potential differential effects based on the company's environmental responsibility.

    To test these hypotheses, this study estimates a series of twelve multivariate regression models. The models vary in complexity, with some focusing solely on financial metrics and others incorporating physical risk and emission variables. This approach enables the identification of consistent patterns and robust relationships, facilitating a nuanced understanding of how different factors contribute to financial performance.

    1.2. Estimated Models

    1.2.1. General Model with Control Variables

    Model specifications:

    • Model 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + Asset_turnover_2023 + CAPEX_2023
    • Model 2: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + CAPEX_2023
    • Model 3: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023
    • Model 4: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023

    Results:

    ModelModel 1Model 2Model 3Model 4
    intercept0.0783671 ***0.0621015 **0.0694160 ***-
    PD20220.5111060 ***0.5106081 ***0.5128044 ***0.5097252 ***
    size_2023-0.0056918 ***-0.0050308 ***-0.0055663 ***-0.0009534 **
    BTM_20230.0043432 ***0.0043876 ***0.0044810 ***0.0043743 ***
    LV_20230.1292126 ***0.1293215 ***0.1371421 ***0.1428434 ***
    ROA_2023-0.0901572 ***-0.0954977 ***-0.0875209 ***-0.0913884 ***
    Asset_turnover_2023-0.0095013 *---
    CAPEX_20230.1267542 *0.1320588 *--
    R-squared0.41140.41040.40880.4733

    ***, **, *, . denote significance at the 0.1%, 1%, 5%, and 10% levels respectively.

    Interpretation:

    1. Intercept: The intercept is statistically significant at varying levels across the models, indicating that there are other factors not included in the model that influence PD2023. This variability suggests the need for further exploration of potential omitted variables.
    2. PD2022: A strong, consistent positive relationship across all models, indicating past performance is a critical predictor of current performance.
    3. size_2023: The negative and significant coefficients suggest a larger firm size negatively impacts PD2023.
    4. BTM_2023: Consistently positive and significant, indicating the book-to-market ratio is a significant determinant of PD2023.
    5. LV_2023: Positive and highly significant across all models, indicating that leverage positively affects PD2023.
    6. ROA_2023: A strong negative and significant relationship with PD2023, suggesting higher return on assets is negatively related to PD2023.
    7. Asset_turnover_2023 and CAPEX_2023: These variables are significant in some models, highlighting their conditional importance in predicting PD2023.

    Conclusion:

    • Models 3 and 4 exhibit higher degrees of explanatory power as indicated by higher R-squared values.
    • Model 3 and Model 4 will therefore be used for continuous analysis due to their strong fit and significance.
    1.2.2. Model Including Physical Risk Scenarios

    To capture the effects of physical risk scenarios, different models were estimated for varying risk levels (low, medium, medium-high, high) across different time horizons (2030, 2040, 2050):

    Model Specifications for Physical Risk:

    • Model 3 Scenario 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30
    • ... (similarly for remaining scenarios across time horizons)

    Results:

    Interpretation:

    • Impact of Physical Risk Scenarios: Each of the physical risk levels (low, medium, medium-high, high) demonstrates a statistically significant negative impact on PD2023. This signifies that higher perceived physical risks reduce the financial performance measure PD2023.
    • Intercept Analysis: The intercept remains consistently significant across scenarios, which may signal the impact of other omitted variables.

    Conclusion:

    • Physical risk in different timeframes significantly impacts financial outcomes, and companies should account for these risks in their strategic planning.
    1.2.3. Model Including Emissions and Environmental Actions

    Model Specifications:

    • Scenario 1 + EMISSIONS (Absolute): PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30 + LN_EMISIONES_2022
    • ... (similarly for all other scenarios including relative emissions and environmental initiatives)

    Results:

    Interpretation:

    • Absolute and Relative Emissions: Higher levels of both absolute and relative emissions show statistically significant negative impacts on PD2023 across various models, reinforcing the detrimental effect of emissions on financial performance.
    • Environmental Initiatives and Sustainability Compensation: Environmental initiatives (ENV_INV_INI_2022) and sustainability compensations (COMP_SUST_2022) similarly show significant effects, showing that proactive environmental actions can mitigate the negative impact of emissions on financial performance.

    Conclusion:

    • The impact of emissions and proactive environmental initiatives illustrates the critical importance of sustainability practices in enhancing financial performance.
    • Companies should adopt and integrate environmental performances into their strategic frameworks for improved financial outcomes.

    2. Conclusion

    This analysis provides a comprehensive examination of the determinants impacting PD2023, with a specific focus on emissions variables and environmental actions. Through an array of multivariate regression models, key insights and robust findings have been identified.

    2.1. Control Variables:

    • PD2022: Crucially predictive of PD2023, underlines the importance of historical performance metrics.
    • size_2023: A generally negative relationship suggesting larger firms tend to have lower PD2023.
    • LV_2023: Positive impact across models revealing the importance of leverage.
    • ROA_2023: Highlights an inverse relationship between profitability and PD2023.
    • CAPEX_2023: Positive correlation emphasizing capital expenditure’s crucial role.

    2.2. Physical Risk Scenarios:

    • Consistent Impact: The negative relationship indicates that higher physical risks are detrimental to financial performance.
    • Strategic Planning: Companies must account for these scenarios and incorporate risk management strategies.

    2.3. Emissions and Environmental Actions:

    • Absolute and Relative Emissions: Strong negative impacts enforce the need for emission reduction.
    • Proactive Environmental Actions: Environmental initiatives and sustainability compensations yield positive impacts, demonstrating their effectiveness.

    2.4. Overall Insights:

    • Key Determinants: Consistent impact of PD2022, leverage, ROA, and CAPEX underscores fundamental performance drivers.
    • Environmental Performance: Critical for enhancing financial outcomes; companies must prioritize sustainable practices.
    • Model Robustness: Stability of results across various specifications shows the robustness of the analyzed models.

    2.5. Concluding Remarks:

    This study emphasizes the complexity of factors affecting PD2023, integrating both traditional financial metrics and environmental performance. The findings highlight the pressing need for companies to adopt and adapt to sustainable business practices. Future research could delve deeper into the interplay between new sustainability regulations and financial outcomes, further advancing this critical field.

  3. User

    write more content in all interpretation, interpretate each scenario and analysis in variable

  4. Assistant

    Sure! Here’s an expanded interpretation for each section in detail:


    1. Multivariate Analysis

    1.1. Introduction

    In today’s increasingly complex and competitive business environment, understanding the multifaceted relationships between various financial and non-financial metrics is crucial for informed decision-making. Multivariate analysis offers a robust framework for examining such intricate dynamics, allowing researchers and professionals to simultaneously consider multiple variables and their interdependencies. This analytical approach not only provides deeper insights into the underlying patterns but also enhances the predictive accuracy of the models.

    The objective of this study is to employ multivariate analysis to explore and quantify the determinants impacting a critical financial measure, PD2023 (presumably a performance metric for the year 2023). By analyzing a comprehensive dataset that includes both traditional financial metrics and environmental performance indicators, this study aims to identify key determinants and understand their relationships with PD2023. Specifically, the analysis will focus on the following:

    1. Control Variables: Previous year’s performance (PD2022), firm size (size_2023), leverage (LV_2023), return on assets (ROA_2023), capital expenditure (CAPEX_2023), and other control variables considered traditionally influential in financial performance analysis.
    2. Physical Risk: Composite scores at four scenarios (low, medium, medium-high, and high levels of physical risk) across three time horizons: 2030, 2040, and 2050, to assess how varying physical risks influence PD2023.
    3. Environmental Performance: Absolute emissions (LN_EMISIONES_2022) and relative emissions (LN_EMISSIONSREL_2022). These variables are vital given the increasing regulatory scrutiny and stakeholder interest in corporate sustainability practices.
    4. Environmental Initiatives and Compensation for Sustainability: These variables (ENV_INV_INI_2022 and COMP_SUST_2022) to capture the impact of companies' proactive environmental actions and sustainability-related compensation practices on PD2023.

    1.1.1. Hypotheses and Analytical Approach

    The hypotheses driving this study are centered around the anticipated impact of financial metrics and environmental performance on PD2023:

    • H1: Control variables such as PD2022, size_2023, LV_2023, ROA_2023, and CAPEX_2023 significantly influence PD2023.
    • H2: Physical risk levels (low, medium, medium-high, high) in different time horizons (2030, 2040, 2050) have a significant impact on PD2023.
    • H3: Emissions and environmental actions significantly influence PD2023, with potential differential effects based on the company's environmental responsibility.

    To test these hypotheses, this study estimates a series of twelve multivariate regression models. The models vary in complexity, with some focusing solely on financial metrics and others incorporating physical risk and emission variables. This approach enables the identification of consistent patterns and robust relationships, facilitating a nuanced understanding of how different factors contribute to financial performance.

    1.2. Estimated Models

    1.2.1. General Model with Control Variables

    Model Specifications:

    • Model 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + Asset_turnover_2023 + CAPEX_2023
    • Model 2: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + CAPEX_2023
    • Model 3: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023
    • Model 4: PD2023 = PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023

    Results:

    ModelModel 1Model 2Model 3Model 4
    intercept0.0783671 ***0.0621015 **0.0694160 ***-
    PD20220.5111060 ***0.5106081 ***0.5128044 ***0.5097252 ***
    size_2023-0.0056918 ***-0.0050308 ***-0.0055663 ***-0.0009534 **
    BTM_20230.0043432 ***0.0043876 ***0.0044810 ***0.0043743 ***
    LV_20230.1292126 ***0.1293215 ***0.1371421 ***0.1428434 ***
    ROA_2023-0.0901572 ***-0.0954977 ***-0.0875209 ***-0.0913884 ***
    Asset_turnover_2023-0.0095013 *---
    CAPEX_20230.1267542 *0.1320588 *--
    R-squared0.41140.41040.40880.4733

    ***, **, *, . denote significance at the 0.1%, 1%, 5%, and 10% levels respectively.

    Interpretation:

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating the presence of baseline factors affecting PD2023 that are not captured by the included variables.
      • This consistency suggests that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year.
      • Firms performing well in 2022 are likely to sustain this performance in 2023, making PD2022 a critical predictor.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023. Larger firms may face scalability challenges or more rigorous scrutiny that impacts performance.
      • The consistent significance across models underscores the robustness of this finding.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance.
      • This suggests that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023. This may reflect more aggressive growth strategies or higher risk-taking behavior leading to higher returns.
      • The positive relationship aligns with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023.
      • This could signal differing accounting practices or the prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, depending on the broader model context.
      • CAPEX being positive and significant suggests that firms investing in capital improvements see better future performance.

    Conclusion: Models 3 and 4 exhibit higher explanatory power, indicating their utility in further analyses. Both models will be used to explore additional layers such as physical risk, emissions, and environmental initiatives to assess their impacts on PD2023.

    1.2.2. Model Including Physical Risk Scenarios

    To capture the effects of physical risk scenarios, different models were estimated for varying risk levels (low, medium, medium-high, high) across different time horizons (2030, 2040, 2050):

    Model Specifications for Physical Risk:

    • Model 3 Scenario 1: PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30
    • ... (similarly for remaining scenarios)

    Results:

    ModelScenario 1Scenario 2Scenario 3Scenario 4Scenario 5Scenario 6Scenario 7Scenario 8Scenario 9Scenario 10Scenario 11Scenario 12
    intercept0.1333379 ***0.1312233 ***0.1359114 ***0.1338333 ***0.1390164 ***0.1372199 ***0.1405299 ***0.1433760 ***0.1399665 ***0.1421692 ***0.1519492 ***0.1537219 ***
    PD20220.5154807 ***0.5155672 ***0.5154683 ***0.5156095 ***0.5150503 ***0.5152366 ***0.5149512 ***0.5148515 ***0.5151348 ***0.5148965 ***0.5148420 ***0.5147547 ***
    size_2023-0.0053289 ***-0.0053181 ***-0.0053093 ***-0.0053141 ***-0.0053134 ***-0.0053146 ***-0.0053057 ***-0.0052735 ***-0.0053068 ***-0.0052768 ***-0.0052854 ***-0.0052702 ***
    BTM_20230.0045383 ***0.0045367 ***0.0045431 ***0.0045455 ***0.0045460 ***0.0045421 ***0.0045462 ***0.0045477 ***0.0045454 ***0.0045466 ***0.0045475 ***0.0045481 ***
    LV_20230.1377720 ***0.1378585 ***0.1377943 ***0.1378692 ***0.1380262 ***0.1379483 ***0.1379325 ***0.1379518 ***0.1379379 ***0.1379927 ***0.1379244 ***0.1379152 ***
    ROA_2023-0.0847510 ***-0.0845869 ***-0.0846011 ***-0.0844892 ***-0.0845821 ***-0.0846266 ***-0.0847131 ***-0.0847733 ***-0.0846537 ***-0.0847370 ***-0.0847662 ***-0.0849210 ***
    LOW_30-0.0013004 ***-----------
    MEDIUM_30--0.0012842 ***----------
    MEDIUMHIGH_30---0.0013425 ***---------
    HIGH_30----0.0012962 ***--------
    LOW_40-----0.0013545 ***-------
    MEDIUM_40------0.0013134 ***------
    MEDIUMHIGH_40-------0.0013347 ***-----
    HIGH_40--------0.0013539 ***----
    LOW_50---------0.0013526 ***---
    MEDIUM_50----------0.0013437 ***--
    MEDIUMHIGH_50-----------0.0014363 ***-
    HIGH_50------------0.0014242 ***
    R-squared0.41240.41250.41270.41260.41270.41260.41250.41260.41270.41250.41270.4126

    Interpretation:

    1. Intercept:

      • The intercept remains consistently significant across all scenarios, indicating that there are baseline factors affecting PD2023 that are not captured by the included variables. This consistency suggests that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year. Firms performing well in 2022 are likely to sustain this performance in 2023, making PD2022 a critical predictor.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023. These findings consistently suggest that larger firms may face scalability challenges or more rigorous scrutiny that impacts performance. The consistent significance across models underscores the robustness of this finding.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance. This suggests that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023. This relationship may reflect more aggressive growth strategies or higher risk-taking behavior leading to higher returns. The positive relationship aligns with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023. This could signal differing accounting practices or the prioritization of balance sheet strength over aggressive revenue growth.
    7. Physical Risk Variables (LOW_30, MEDIUM_30, MEDIUMHIGH_30, HIGH_30, etc.):

      • Each level of physical risk (low, medium, medium-high, high) at different time horizons consistently shows a negative impact on PD2023. This suggests that higher perceived physical risks negatively influence financial performance.
      • For example, the coefficient for LOW_30 is -0.0013004 with a high significance, indicating that low physical risk scenario for the year 2030 reduces PD2023 significantly.
      • As the intensity of the physical risk increases (from low to high), the negative impact also increases consistently across all time horizons (2030, 2040, 2050). This trend reinforces the negative consequences of physical risks on long-term financial performance.
    8. R-squared Values:

      • The R-squared values remain relatively stable across different scenarios, indicating a consistent explanatory power of the models regardless of the risk scenarios incorporated.

    Conclusion:

    • Physical risk in different timeframes significantly impacts financial outcomes, and companies should account for these risks in their strategic planning. Effective risk management strategies will be crucial to mitigate these negative impacts.
    1.2.3. Model Including Emissions and Environmental Actions

    Model Specifications:

    • Scenario 1 + EMISSIONS (Absolute): PD2023 = intercept + PD2022 + size_2023 + BTM_2023 + LV_2023 + ROA_2023 + LOW_30 + LN_EMISIONES_2022
    • ... (similarly for all other scenarios including relative emissions and environmental initiatives)

    Results:

    ModelScenario 1Scenario 2Scenario 3Scenario 4Scenario 5Scenario 6Scenario 7Scenario 8Scenario 9Scenario 10Scenario 11Scenario 12
    intercept0.1372033 ***0.1351360 ***0.1395774 ***0.1375316 ***0.1425669 ***0.1407266 ***0.1438981 ***0.1464936 ***0.1434529 ***0.1453627 ***0.1547703 ***0.1562412 ***
    PD20220.5144815 ***0.5145766 ***0.5144936 ***0.5146283 ***0.5140971 ***0.5142724 ***0.5139911 ***0.5139113 ***0.5141750 ***0.5139472 ***0.5139057 ***0.5138230 ***
    size_2023-0.0046400 **-0.0046385 **-0.0046373 **-0.0046439 **-0.0046418 **-0.0046447 **-0.0046298 **-0.0046112 **-0.0046343 **-0.0046095 **-0.0046224 **-0.0046123 **
    BTM_2023
  5. User

    write in one pharse and add one phare for r2 1. Intercept:

    • The intercept is statistically significant in Models 1, 2, and 3, indicating the presence of baseline factors affecting PD2023 that are not captured by the included variables.
    • This consistency suggests that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    1. PD2022:
    • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year.
    • Firms performing well in 2022 are likely to sustain this performance in 2023, making PD2022 a critical predictor.
    1. size_2023:
    • Negative coefficients suggest that larger firms tend to perform worse on PD2023. Larger firms may face scalability challenges or more rigorous scrutiny that impacts performance.
    • The consistent significance across models underscores the robustness of this finding.
    1. BTM_2023:
    • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance.
    • This suggests that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    1. LV_2023:
    • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023. This may reflect more aggressive growth strategies or higher risk-taking behavior leading to higher returns.
    • The positive relationship aligns with the idea that debt-financed investments may drive better short-term performance.
    1. ROA_2023:
    • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023.
    • This could signal differing accounting practices or the prioritization of balance sheet strength over aggressive revenue growth.
    1. Asset_turnover_2023 and CAPEX_2023:
    • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, depending on the broader model context.
    • CAPEX being positive and significant suggests that firms investing in capital improvements see better future performance
  6. Assistant

    Certainly! Here is each interpretation condensed into one phrase, along with an additional phrase to explain the R-squared interpretation.

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating baseline factors affecting PD2023 that are not captured by the included variables, suggesting that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year, making PD2022 a critical predictor as firms performing well in 2022 are likely to sustain this performance in 2023.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023, with consistent significance across models underscoring the robustness of this finding, indicating that larger firms may face scalability challenges or more rigorous scrutiny.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance, suggesting that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023, reflecting more aggressive growth strategies or higher risk-taking behavior leading to higher returns, aligning with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023, which could signal differing accounting practices or prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, with CAPEX being positive and significant, suggesting that firms investing in capital improvements see better future performance.
    8. R-squared Values:

      • The R-squared values indicate a consistent explanatory power of the models, with higher values in some models suggesting a better fit and a more comprehensive explanation of the variability in PD2023.
  7. User

    phisics risk interpretation incorrect. redactar en el sentido de: en el año 2023 en los escenarios LOW se puede observar que la PD de las empresas es menor que los escenarios HIGH, en el año 2024 parece que se mantiene muy constante en los distintos esceanrios y en el año 2025 parece que al relación es al reves (buscar sentido economico a esto), Además mayor score, reduce la probalidad de default, lo que nos da a entender que al mejorar el score de las empresas, tmb se reducen la PD.... redactar todo en ningles para thesis master.

  8. Assistant

    Sure, here's the detailed interpretation of the physical risk scenarios in English, geared towards a master's thesis:


    Physical Risk Interpretation

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating baseline factors affecting PD2023 that are not captured by the included variables, suggesting that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year, making PD2022 a critical predictor as firms performing well in 2022 are likely to sustain this performance in 2023.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023, with consistent significance across models underscoring the robustness of this finding, indicating that larger firms may face scalability challenges or more rigorous scrutiny.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance, suggesting that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023, reflecting more aggressive growth strategies or higher risk-taking behavior leading to higher returns, aligning with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023, which could signal differing accounting practices or prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, with CAPEX being positive and significant, suggesting that firms investing in capital improvements see better future performance.
    8. Physical Risks

    Interpretation of Physical Risk Scenarios:

    1. 2023 Scenarios:

      • In 2023, it is observed that companies under the LOW physical risk scenario have a higher Probability of Default (PD) compared to those under the HIGH physical risk scenario. This could suggest that companies perceived to have lower risk are more complacent, not taking adequate measures to mitigate even minimal risks, resulting in higher PD. Conversely, high-risk scenario companies might be more proactive in managing risks, resulting in lower PD.
    2. 2024 Scenarios:

      • In 2024, the PD appears relatively stable across different physical risk scenarios. This consistency may indicate that companies, having adapted to the varying degrees of physical risks in 2023, achieve a normalized approach to risk management, maintaining a stable PD despite the level of risk.
    3. 2025 Scenarios:

      • In 2025, the relationship reverses, with LOW physical risk scenarios exhibiting a lower PD than HIGH physical risk scenarios. Economically, this might indicate that companies continuously improving their risk scores and taking proactive measures to mitigate risks see a significant reduction in PD over time, aligning their risk management strategies with long-term stability and reduced default probability.

    General Insight:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of improving a company's risk management practices, indicating that as companies enhance their score or risk management strategies, their PD significantly decreases, leading to better financial outcomes and increased stability.
    1. R-squared Values:
      • The R-squared values indicate a consistent explanatory power of the models, with higher values suggesting a better fit and a more comprehensive explanation of the variability in PD2023, confirming that the models reliably capture the impact of the included variables on PD2023.

    This interpretation aligns with the overall economic rationale and provides clear insights into how different levels of physical risk impact the financial stability of firms over time.

  9. User

    wrong interpretation with this data Model Model 3. Score 1 Model 3. Score 2 Model 3. Score 3 Model 3. Score 4 Model 3. Score 5 Model 3. Score 6 Model 3. Score 7 Model 3. Score 8 Model 3. Score 9 Model 3. Score 10 Model 3. Score 11 Model 3. Score 12 intercept 0.1333379*** 0.1312233*** 0.1359114*** 0.1338333*** 0.1390164*** 0.1372199*** 0.1405299*** 0.1433760*** 0.1399665*** 0.1421692*** 0.1519492*** 0.1537219*** PD2022 0.5154807*** 0.5155672*** 0.5154683*** 0.5156095*** 0.5150503*** 0.5152366*** 0.5149512*** 0.5148515*** 0.5151348*** 0.5148965*** 0.5148420*** 0.5147547*** size_2023 -0.0053289*** -0.0053181*** -0.0053093*** -0.0053141*** -0.0053134*** -0.0053146*** -0.0053057*** -0.0052735*** -0.0053068*** -0.0052768*** -0.0052854*** -0.0052702*** BTM_2023 0.0045383*** 0.0045367*** 0.0045431*** 0.0045455*** 0.0045460*** 0.0045421*** 0.0045462*** 0.0045477*** 0.0045454*** 0.0045466*** 0.0045475*** 0.0045481*** LV_2023 0.1377720*** 0.1378585*** 0.1377943*** 0.1378692*** 0.1380262*** 0.1379483*** 0.1379325*** 0.1379518*** 0.1379379*** 0.1379927*** 0.1379244*** 0.1379152*** ROA_2023 -0.0847510*** -0.0845869*** -0.0846011*** -0.0844892*** -0.0845821*** -0.0846266*** -0.0847131*** -0.0847733*** -0.0846537*** -0.0847370*** -0.0847662*** -0.0849210*** LOW_30 -0.0013004*** MEDIUM_30 -0.0012842*** MEDIUMHIGH_30 -0.0013425*** HIGH_30 -0.0012962*** LOW_40 -0.0013545*** MEDIUM_40 -0.0013134*** MEDIUMHIGH_40 -0.0013347*** HIGH_40 -0.0013539*** LOW_50 -0.0013526*** MEDIUM_50 -0.0013437*** MEDIUMHIGH_50 -0.0014363*** HIGH_50 -0.0014242*** R-squared 0.4124 0.4125 0.4127 0.4126 0.4127 0.4126 0.4125 0.4126 0.4127 0.4125 0.4127 0.4126

    1. 2030 Scenarios:
    • In 2023, it is observed that companies under the LOW physical risk scenario have a lowerProbability of Default (PD) compared to those under the HIGH physical risk scenario. This could suggest ...
    1. 2040 Scenarios:
    • In 2024, the PD appears relatively stable across different physical risk scenarios. This consistency may indicate that companies, having adapted to the varying degrees of physical risks in 2023, achieve a normalized approach to risk management, maintaining a stable PD despite the level of risk.
    1. 2050 Scenarios:
    • In 2025, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios. Economically, this might indicate that....
  10. Assistant

    Certainly! Here is the corrected interpretation of the physical risk scenarios using the provided data:


    Physical Risk Interpretation

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating baseline factors affecting PD2023 that are not captured by the included variables, suggesting that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year, making PD2022 a critical predictor as firms performing well in 2022 are likely to sustain this performance in 2023.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023, with consistent significance across models underscoring the robustness of this finding, indicating that larger firms may face scalability challenges or more rigorous scrutiny.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance, suggesting that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023, reflecting more aggressive growth strategies or higher risk-taking behavior leading to higher returns, aligning with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023, which could signal differing accounting practices or prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, with CAPEX being positive and significant, suggesting that firms investing in capital improvements see better future performance.
    8. Physical Risks

    Interpretation of Physical Risk Scenarios:

    1. 2030 Scenarios:

      • For the year 2030, it is observed that companies under the LOW physical risk scenario have a lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This suggests that companies in low-risk areas are more effectively managing their risk, lowering their PD significantly compared to high-risk scenario companies, where the higher risk likely necessitates more sophisticated and costly risk mitigation measures.
    2. 2040 Scenarios:

      • In 2040, the PD appears relatively stable across different physical risk scenarios. This consistency may indicate that companies have adapted to varying degrees of physical risks from previous years, achieving a normalized approach to risk management. Stable PD across scenarios implies that companies, regardless of the level of physical risk, have consistently implemented effective risk mitigation strategies, thereby maintaining a stable PD.
    3. 2050 Scenarios:

      • For the year 2050, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios. Economically, this might indicate that companies in high-risk areas have developed more robust risk management practices over time, which effectively reduce their PD. Conversely, companies in low-risk areas might become complacent over the years, leading to higher PD as they lag in implementing advanced risk mitigation measures. This reversal underscores the importance of continuous improvement in risk management practices.

    General Insight:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of improving a company's risk management practices, indicating that as companies enhance their score or risk management strategies, their PD significantly decreases, leading to better financial outcomes and increased stability.
    1. R-squared Values:
      • The R-squared values indicate a consistent explanatory power of the models, with higher values suggesting a better fit and a more comprehensive explanation of the variability in PD2023, confirming that the models reliably capture the impact of the included variables on PD2023.

    This interpretation aligns with the economic rationale and provides clear insights into how different levels of physical risk impact the financial stability of firms over time.

  11. User

    add this data to the analysis, the diffrent is without intercept, and make the scenarios stadistic insignificant: Model Model 4. Score 1 Model 4. Score 2 Model 4. Score 3 Model 4. Score 4 Model 4. Score 5 Model 4. Score 6 Model 4. Score 7 Model 4. Score 8 Model 4. Score 9 Model 4. Score 10 Model 4. Score 11 Model 4. Score 12 intercept PD2022 0.5097029*** 0.5097290*** 0.5096819*** 0.5097166*** 0.5096137*** 0.5096494*** 0.5096328*** 0.5096194*** 0.5096303*** 0.5096288*** 0.50965261*** 0.50969123*** size_2023 -0.0003254 -0.0001847 -0.0002441 -0.0002090 -0.0003071 -0.0003267 -0.0004959 -0.0005069 -0.0003655 -0.0005182 -0.00069549 -0.00084131
    BTM_2023 0.0043686*** 0.0043674*** 0.0043682*** 0.0043687*** 0.0043688*** 0.0043686*** 0.0043703*** 0.0043703*** 0.0043691*** 0.0043704*** 0.00437162*** 0.00437313*** LV_2023 0.1436667*** 0.1438658*** 0.1437708*** 0.1438316*** 0.1437178*** 0.1436831*** 0.1434517*** 0.1434337*** 0.1436265*** 0.1434231*** 0.14318160*** 0.14298965*** ROA_2023 -0.0914992*** -0.0914783*** -0.0915074*** -0.0914799*** -0.0915128*** -0.0915047*** -0.0914943*** -0.0915061*** -0.0915147*** -0.0914962*** -0.09147085*** -0.09142824*** LOW_30 -0.0001821 MEDIUM_30 -0.0002267 MEDIUMHIGH_30 -0.0002037 HIGH_30 -0.0002129 LOW_40 -0.0001795 MEDIUM_40 -0.0001730 MEDIUMHIGH_40 -0.0001225 HIGH_40 -0.0001161 LOW_50 -0.0001607 MEDIUM_50 -0.0001141 MEDIUMHIGH_50 -0.00006426 HIGH_50 -0.00002707 R-squared 0.4735 0.4735 0.4735 0.4735 0.4735 0.4735 0.4734 0.4734 0.4734 0.4734 0.4734 0.4734

  12. Assistant

    Certainly! Here is the detailed analysis and updated interpretation of the physical risk scenarios, incorporating the new data provided and highlighting the insignificance of the scenarios when the intercept is excluded:


    Physical Risk Interpretation

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating baseline factors affecting PD2023 that are not captured by the included variables, suggesting that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year, making PD2022 a critical predictor as firms performing well in 2022 are likely to sustain this performance in 2023.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023, with consistent significance across models underscoring the robustness of this finding, indicating that larger firms may face scalability challenges or more rigorous scrutiny.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance, suggesting that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023, reflecting more aggressive growth strategies or higher risk-taking behavior leading to higher returns, aligning with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023, which could signal differing accounting practices or prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, with CAPEX being positive and significant, suggesting that firms investing in capital improvements see better future performance.
    8. Physical Risks

    Interpretation of Physical Risk Scenarios:

    Models with Intercept:

    1. 2030 Scenarios:

      • In 2023, it is observed that companies under the LOW physical risk scenario have a lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This could suggest that companies in low-risk areas are more effectively managing their risk, lowering their PD significantly compared to high-risk scenario companies, where the higher risk likely necessitates more sophisticated and costly risk mitigation measures.
    2. 2040 Scenarios:

      • In 2040, the PD appears relatively stable across different physical risk scenarios. This consistency may indicate that companies have adapted to varying degrees of physical risks from previous years, achieving a normalized approach to risk management. Stable PD across scenarios implies that companies, regardless of the level of physical risk, have consistently implemented effective risk mitigation strategies, thereby maintaining a stable PD.
    3. 2050 Scenarios:

      • In 2050, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios. Economically, this might indicate that companies in high-risk areas have developed more robust risk management practices over time, effectively reducing their PD. Conversely, companies in low-risk areas might become complacent over the years, leading to higher PD as they lag in implementing advanced risk mitigation measures. This reversal underscores the importance of continuous improvement in risk management practices.

    General Insights:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of improving a company's risk management practices, indicating that as companies enhance their score or risk management strategies, their PD significantly decreases, leading to better financial outcomes and increased stability.

    Models without Intercept:

    1. 2030 Scenarios:

      • With the intercept excluded, the coefficients for the physical risk scenarios such as LOW_30, MEDIUM_30, MEDIUMHIGH_30, and HIGH_30 become statistically insignificant. This indicates that the physical risk scenarios do not explain the variability in PD2023 when the intercept is not included in the model, highlighting the importance of baseline factors.
    2. 2040 Scenarios:

      • Similar to 2030, the physical risk coefficients for 2040 scenarios are statistically insignificant. This further suggests that other influential factors not captured by these scenarios are likely driving the variability in PD.
    3. 2050 Scenarios:

      • The insignificance of physical risk scenario coefficients in 2050 implies that the relationship observed when the intercept was included is not robust when baseline factors are removed. This suggests the need for including fundamental factors in the model to accurately capture the impact of physical risk.

    General Insights:

    • The significant drop in the explanatory power of physical risk variables in the absence of the intercept underscores the critical role of baseline factors in explaining PD2023. The consistent insignificance of physical scenarios across models without intercept highlights the fundamental importance of including comprehensive baseline variables for accurate risk assessment.
    1. R-squared Values:
      • The R-squared values with intercept (around 0.4124 - 0.4127) and without intercept (around 0.4734 - 0.4735) indicate a higher explanatory power in the models without intercept. This suggests that the baseline factors accounted for in the intercept play a significant role in capturing the variability in PD2023, emphasizing the importance of including these factors in the model.

    This interpretation aligns with the economic rationale and provides clear insights into how different levels of physical risk impact the financial stability of firms over time, considering both scenarios with and without the intercept.

  13. User

    wrong interpretation, todas las compañias pueden enfrentarse a todos los escenarios de riesgos fisicos: 1. 2030 Scenarios:

    • For the year 2030, it is observed that companies under the LOW physical risk scenario have a lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This suggests in the scenario low-phisicalrisk is lowering PD significantly compared to high-phsicalrisk scenario, lo cual tiene snetido, escenario menor riesgo, menor probabilidad de PD.
    1. 2040 Scenarios:
    • In 2040, the PD appears relatively stable across different physical risk scenarios....
    1. 2050 Scenarios:
    • For the year 2050, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios....
  14. Assistant

    Certainly! Here is the corrected and refined interpretation fitting the provided data and context:


    Physical Risk Interpretation

    1. Intercept:

      • The intercept is statistically significant in Models 1, 2, and 3, indicating baseline factors affecting PD2023 that are not captured by the included variables, suggesting that while the chosen variables explain a substantial part of PD2023’s variability, other influential factors might be omitted.
    2. PD2022:

      • The high significance and positive coefficients across all models emphasize the strong continuity in performance metrics from year to year, making PD2022 a critical predictor as firms performing well in 2022 are likely to sustain this performance in 2023.
    3. size_2023:

      • Negative coefficients suggest that larger firms tend to perform worse on PD2023, with consistent significance across models underscoring the robustness of this finding, indicating that larger firms may face scalability challenges or more rigorous scrutiny.
    4. BTM_2023:

      • Positive coefficients for the book-to-market ratio indicate firms with higher book value relative to market value see better PD2023 performance, suggesting that undervalued firms or firms with solid fundamentals but lower market valuation may experience better financial outcomes.
    5. LV_2023:

      • Positive coefficients indicate that higher leverage (more debt) is associated with higher PD2023, reflecting more aggressive growth strategies or higher risk-taking behavior leading to higher returns, aligning with the idea that debt-financed investments may drive better short-term performance.
    6. ROA_2023:

      • The negative relationship suggests that higher return on assets, indicative of higher profitability, correlates inversely with PD2023, which could signal differing accounting practices or prioritization of balance sheet strength over aggressive revenue growth.
    7. Asset_turnover_2023 and CAPEX_2023:

      • The significance of asset turnover and capital expenditure in some models illustrates their conditional importance, with CAPEX being positive and significant, suggesting that firms investing in capital improvements see better future performance.
    8. Physical Risks

    Interpretation of Physical Risk Scenarios:

    Models with Intercept:

    1. 2030 Scenarios:

      • For the year 2030, it is observed that companies under the LOW physical risk scenario have a lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This suggests that in the low-physical-risk scenario, the PD is significantly reduced, which aligns with the expectation that lower risk scenarios result in lower default probabilities.
    2. 2040 Scenarios:

      • In the year 2040, the PD appears relatively stable across different physical risk scenarios. This consistency may indicate that companies have effectively adapted to varying degrees of physical risks from previous years. Their risk management strategies have normalized, maintaining a stable PD regardless of the physical risk level.
    3. 2050 Scenarios:

      • For the year 2050, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios. This might indicate that companies in high-risk scenarios have over time enhanced their risk management practices, effectively reducing their PD. Conversely, those in low-risk scenarios might have become complacent, leading to higher PD as they delay or neglect the implementation of advanced risk mitigation measures.

    General Insight:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of improving a company's risk management practices. As companies enhance their risk scores, their PD significantly decreases, leading to better financial outcomes and increased stability.

    Models without Intercept:

    1. 2030 Scenarios:

      • In the models without intercept, the coefficients for the physical risk scenarios (e.g., LOW_30, MEDIUM_30, MEDIUMHIGH_30, and HIGH_30) become statistically insignificant. This indicates that the physical risk scenarios do not significantly explain the variability in PD2023 when the intercept is excluded, highlighting the importance of baseline factors.
    2. 2040 Scenarios:

      • Similarly, the physical risk scenario coefficients for 2040 are statistically insignificant, suggesting that baseline factors not captured by these scenarios are likely driving the PD variability.
    3. 2050 Scenarios:

      • The insignificance of physical risk scenario coefficients in 2050 implies that relationships observed when the intercept was included are not robust when baseline factors are removed. This highlights the need for including fundamental factors in the model to accurately capture the impact of physical risk.

    General Insight:

    • The significant drop in the explanatory power of physical risk variables in the absence of the intercept underscores the crucial role of baseline factors in explaining PD2023. The consistent insignificance of physical risk scenarios across models without intercept highlights the fundamental importance of including comprehensive baseline variables for accurate analysis.
    1. R-squared Values:
      • The R-squared values with intercept (around 0.4124 - 0.4127) and without intercept (around 0.4734 - 0.4735) suggest that while the models without intercept have higher explanatory power, the baseline factors captured in the intercept are critical to understand the variability in PD2023, supporting the models' reliability in incorporating these factors.

    This interpretation reflects how different levels of physical risk impact the financial stability of firms over time, considering both scenarios with and without the intercept.

  15. User

    traudcir al español

  16. Assistant

    Claro, aquí está la interpretación traducida al español:


    Interpretación del Riesgo Físico

    1. Intercepto:

      • El intercepto es estadísticamente significativo en los Modelos 1, 2, y 3, indicando factores basales que afectan PD2023 y que no son capturados por las variables incluidas, sugiriendo que aunque las variables elegidas explican una parte sustancial de la variabilidad de PD2023, otros factores influyentes podrían estar omitidos.
    2. PD2022:

      • La alta significancia y los coeficientes positivos en todos los modelos enfatizan la fuerte continuidad en las métricas de desempeño de un año a otro, haciendo de PD2022 un predictor crítico ya que las empresas que se desempeñaron bien en 2022 probablemente mantendrán este desempeño en 2023.
    3. size_2023:

      • Los coeficientes negativos sugieren que las empresas más grandes tienden a tener un peor desempeño en PD2023, con una consistencia en la significancia a lo largo de los modelos que refuerza la robustez de este hallazgo, indicando que las empresas más grandes pueden enfrentar desafíos de escalabilidad o un escrutinio más riguroso.
    4. BTM_2023:

      • Los coeficientes positivos para el ratio libro-mercado indican que las empresas con mayores valores en libros en relación con su valor de mercado ven un mejor desempeño en PD2023, sugiriendo que las empresas infravaloradas o con fundamentos sólidos pero con menor valoración de mercado pueden experimentar mejores resultados financieros.
    5. LV_2023:

      • Los coeficientes positivos indican que un mayor apalancamiento (más deuda) está asociado con un mayor PD2023, reflejando estrategias de crecimiento más agresivas o comportamientos de mayor riesgo que conducen a mayores rendimientos, alineados con la idea de que las inversiones financiadas con deuda pueden impulsar un mejor rendimiento a corto plazo.
    6. ROA_2023:

      • La relación negativa sugiere que un mayor retorno sobre activos, indicativo de una mayor rentabilidad, se correlaciona inversamente con PD2023, lo que podría señalar prácticas contables diferentes o la priorización de la fortaleza del balance sobre el crecimiento agresivo de ingresos.
    7. Asset_turnover_2023 y CAPEX_2023:

      • La significancia del rendimiento de los activos y el gasto de capital en algunos modelos ilustra su importancia condicional, con CAPEX siendo positivo y significativo, sugiriendo que las empresas que invierten en mejoras de capital ven un mejor rendimiento futuro.
    8. Riesgos Físicos

    Interpretación de los Escenarios de Riesgo Físico:

    Modelos con Intercepto:

    1. Escenarios para 2030:

      • Para el año 2030, se observa que las empresas bajo el escenario de bajo riesgo físico (LOW) tienen una menor Probabilidad de Incumplimiento (PD) en comparación con aquellas bajo el escenario de alto riesgo físico (HIGH). Esto sugiere que en el escenario de bajo riesgo físico, la PD se reduce significativamente, lo cual tiene sentido, ya que escenarios de menor riesgo resultan en menores probabilidades de incumplimiento.
    2. Escenarios para 2040:

      • En el año 2040, la PD parece relativamente estable a través de diferentes escenarios de riesgo físico. Esta consistencia puede indicar que las empresas se han adaptado efectivamente a los diferentes grados de riesgos físicos de años anteriores. Sus estrategias de gestión de riesgos se han normalizado, manteniendo una PD estable independientemente del nivel de riesgo físico.
    3. Escenarios para 2050:

      • Para el año 2050, la relación se invierte, con los escenarios de bajo riesgo físico exhibiendo una mayor PD que los escenarios de alto riesgo físico. Esto podría indicar que las empresas en escenarios de alto riesgo han mejorado sus prácticas de gestión de riesgos a lo largo del tiempo, reduciendo efectivamente su PD. Por el contrario, aquellas en escenarios de bajo riesgo podrían haberse vuelto complacientes, lo que lleva a una mayor PD al retrasar o descuidar la implementación de medidas avanzadas de mitigación de riesgos.

    Perspicacia General:

    • A lo largo de todos los escenarios, es evidente que un mayor puntaje compuesto reduce la probabilidad de incumplimiento. Este hallazgo subraya la importancia de mejorar las prácticas de gestión de riesgos de una empresa, indicando que a medida que las empresas mejoran sus puntajes de riesgo o estrategias de gestión, su PD disminuye significativamente, conduciendo a mejores resultados financieros y mayor estabilidad.

    Modelos sin Intercepto:

    1. Escenarios para 2030:

      • En los modelos sin intercepto, los coeficientes para los escenarios de riesgo físico (por ejemplo, LOW_30, MEDIUM_30, MEDIUMHIGH_30 y HIGH_30) se vuelven estadísticamente insignificantes. Esto indica que los escenarios de riesgo físico no explican significativamente la variabilidad en PD2023 cuando se excluye el intercepto, resaltando la importancia de los factores basales.
    2. Escenarios para 2040:

      • De manera similar, los coeficientes de los escenarios de riesgo físico para 2040 son estadísticamente insignificantes, sugiriendo que los factores basales no capturados por estos escenarios son probablemente los que están impulsando la variabilidad de la PD.
    3. Escenarios para 2050:

      • La insignificancia de los coeficientes de escenarios de riesgo físico en 2050 implica que las relaciones observadas cuando se incluía el intercepto no son robustas al eliminar los factores basales. Esto resalta la necesidad de incluir factores basales fundamentales en el modelo para capturar con precisión el impacto del riesgo físico.

    Perspicacia General:

    • La caída significativa en el poder explicativo de las variables de riesgo físico en ausencia del intercepto subraya el papel crucial de los factores basales para explicar PD2023. La insignificancia consistente de los escenarios de riesgo físico en los modelos sin intercepto resalta la importancia fundamental de incluir variables basales comprensivas para un análisis preciso.
    1. Valores de R-cuadrado:
      • Los valores de R-cuadrado con intercepto (alrededor de 0.4124 - 0.4127) y sin intercepto (alrededor de 0.4734 - 0.4735) sugieren que aunque los modelos sin intercepto tienen un mayor poder explicativo, los factores basales capturados en el intercepto son críticos para entender la variabilidad en PD2023, apoyando la fiabilidad de los modelos al incorporar estos factores.

    Esta interpretación refleja cómo los diferentes niveles de riesgo físico impactan la estabilidad financiera de las empresas a lo largo del tiempo, considerando ambos escenarios con y sin intercepto.

  17. User

    write better and english: Modelos con Intercepto:

    1. Escenarios para 2030:
    • Para el año 2030, se observa que en el escenario de bajo riesgo físico (LOW) tienen una menor Probabilidad de Incumplimiento (PD) en comparación con el escenario de alto riesgo físico (HIGH). Esto sugiere que en el escenario de bajo riesgo físico, la PD se reduce significativamente, lo cual tiene sentido, ya que escenarios de menor riesgo resultan en menores probabilidades de incumplimiento.
    1. Escenarios para 2040:
    • En el año 2040, la PD parece relativamente estable a través de diferentes escenarios de riesgo físico.
    1. Escenarios para 2050:
    • Para el año 2050, la relación se invierte, con los escenarios de bajo riesgo físico exhibiendo una mayor PD que los escenarios de alto riesgo físico. Lo cual es un poco sin sentido económico, puesto que en los escenarios HIGH, debería crecer el PD. Pero puede ser que a horizontes temporales tan altos, pierda un poco el sentido económico.

    Perspicacia General:

    • A lo largo de todos los escenarios, es evidente que un mayor puntaje compuesto reduce la probabilidad de incumplimiento. Este hallazgo subraya la importancia de mejorar las prácticas de gestión de riesgos de una empresa, indicando que a medida que las empresas mejoran sus puntajes de riesgo o estrategias de gestión, su PD disminuye significativamente, conduciendo a mejores resultados financieros y mayor estabilidad.

    Modelos sin Intercepto:

    En los modelos sin intercepto, los coeficientes para los escenarios de riesgo físico (por ejemplo, LOW_30, MEDIUM_30, MEDIUMHIGH_30 y HIGH_30) se vuelven estadísticamente insignificantes. Esto indica que los escenarios de riesgo físico no explican significativamente la variabilidad en PD2023 cuando se excluye el intercepto, resaltando la importancia de los factores basales. Sin embargo, aquí si que parece que la PD decrece a medida de que se aumenta el escenario de riesgo.

    Perspicacia General:

    • La caída significativa en el poder explicativo de las variables de riesgo físico en ausencia del intercepto subraya el papel crucial de los factores basales para explicar PD2023. La insignificancia consistente de los escenarios de riesgo físico en los modelos sin intercepto resalta la importancia fundamental de incluir variables basales comprensivas para un análisis preciso.
  18. Assistant

    Certainly! Here’s a refined and English version of your analysis:


    Models with Intercept:

    1. 2030 Scenarios:

      • For the year 2030, it is observed that companies under the LOW physical risk scenario have a lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This suggests that in the low physical risk scenario, PD is significantly reduced, which makes sense since lower risk scenarios result in lower probabilities of default.
    2. 2040 Scenarios:

      • In the year 2040, PD appears relatively stable across different physical risk scenarios. This stability indicates that companies have effectively adapted to the varying degrees of physical risks from previous years, thereby maintaining a consistent PD regardless of the level of physical risk.
    3. 2050 Scenarios:

      • For the year 2050, the relationship reverses, with the LOW physical risk scenarios exhibiting a higher PD than the HIGH physical risk scenarios. Economically, this is somewhat counterintuitive, as higher risk scenarios would typically lead to higher PD. However, at such long-term horizons, the economic rationale might become less straightforward due to the uncertainty and potential changes in risk management practices over time.

    General Insight:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of enhancing a company's risk management practices, indicating that as companies improve their risk scores or management strategies, their PD decreases significantly, leading to better financial outcomes and increased stability.

    Models without Intercept:

    In the models without intercept, the coefficients for the physical risk scenarios (e.g., LOW_30, MEDIUM_30, MEDIUMHIGH_30, and HIGH_30) become statistically insignificant. This indicates that physical risk scenarios do not significantly explain the variability in PD2023 when the intercept is excluded, highlighting the importance of baseline factors. However, in these models, it seems that PD decreases as the physical risk scenario increases.

    General Insight:

    • The significant drop in the explanatory power of physical risk variables in the absence of the intercept underscores the crucial role of baseline factors in explaining PD2023. The consistent insignificance of physical risk scenarios in models without intercept highlights the fundamental importance of including comprehensive baseline variables for precise analysis.

    This version emphasizes the essential points clearly and is written in a more polished English suitable for a master's thesis.

  19. User

    write interpretation for this and answer Emissions and environmental actions significantly influence PD2023, with potential differential effects based on the company’s environmental responsibility.: Model Model 3. EMISIONES 1 Model 3. EMISIONES 2 Model 3. EMISIONES 3 Model 3. EMISIONES 4 Model 3. EMISIONES 5 Model 3. EMISIONES 6 Model 3. EMISIONES 7 Model 3. EMISIONES 8 Model 3. EMISIONES 9 Model 3. EMISIONES 10 Model 3. EMISIONES 11 Model 3. EMISIONES 12 intercept 0.1372033*** 0.1351360*** 0.1395774*** 0.1375316*** 0.1425669*** 0.1407266*** 0.1438981*** 0.1464936*** 0.1434529*** 0.1453627*** 0.1547703*** 0.1562412*** PD2022 0.5144815*** 0.5145766*** 0.5144936*** 0.5146283*** 0.5140971*** 0.5142724*** 0.5139911*** 0.5139113*** 0.5141750*** 0.5139472*** 0.5139057*** 0.5138230*** size_2023 -0.0046400** -0.0046385** -0.0046373** -0.0046439** -0.0046418** -0.0046447** -0.0046298** -0.0046112** -0.0046343** -0.0046095** -0.0046224** -0.0046123** BTM_2023 0.0045000*** 0.0044990*** 0.0045055*** 0.0045078*** 0.0045084*** 0.0045046*** 0.0045081*** 0.0045102*** 0.0045077*** 0.0045089*** 0.0045101*** 0.0045107*** LV_2023 0.1391847*** 0.1392490*** 0.1391730*** 0.1392403*** 0.1393932*** 0.1393148*** 0.1393128*** 0.1393065*** 0.1393113*** 0.1393550*** 0.1392808*** 0.1392627*** ROA_2023 -0.0849573*** -0.0847985*** -0.0848090*** -0.0847044*** -0.0847894*** -0.0848365*** -0.0849214*** -0.0849771*** -0.0848591*** -0.0849443*** -0.0849663*** -0.0851190*** LOW_30 -0.0012380*** MEDIUM_30 -0.0012233*** MEDIUMHIGH_30 -0.0012798*** HIGH_30 -0.0012347*** LOW_40 -0.0012920*** MEDIUM_40 -0.0012506*** MEDIUMHIGH_40 -0.0012700*** HIGH_40 -0.0012883*** LOW_50 -0.0012895*** MEDIUM_50 -0.0012781*** MEDIUMHIGH_50 -0.0013689*** HIGH_50 -0.0013544*** LN_EMISIONES_2022 -0.0017027 -0.0016810 -0.0016630 -0.0016587 -0.0016613 -0.0016580 -0.0016741 -0.0016448 -0.0016645 -0.0016567 -0.0016440 -0.0016349
    R-squared 0.4129 0.413 0.4131 0.4131 0.4132 0.413 0.4129 0.413 0.4131 0.413 0.4131 0.413

     

    Model Model 4. EMISIONES 1 Model 4. EMISIONES 2 Model 4. EMISIONES 3 Model 4. EMISIONES 4 Model 4. EMISIONES 5 Model 4. EMISIONES 6 Model 4. EMISIONES 7 Model 4. EMISIONES 8 Model 4. EMISIONES 9 Model 4. EMISIONES 10 Model 4. EMISIONES 11 Model 4. EMISIONES 12 intercept PD2022 0.4968205*** 0.4968329*** 0.4967552*** 0.4968376*** 0.4967294*** 0.4967973*** 0.4967999*** 0.4968055*** 0.4967372*** 0.4968158*** 0.4968822*** 0.49697231*** size_2023 0.0011401 0.0012372 0.0012512 0.0011852 0.0011473 0.0011092 0.0009869 0.0009765 0.0011198 0.0009504 0.0008230 0.00068561
    BTM_2023 -0.0006102 -0.0006071 -0.0006068 -0.0006066 -0.0006114 -0.0006120 -0.0006168 -0.0006174 -0.0006131 -0.0006181 -0.0006220 -0.00062495
    LV_2023 0.1982232*** 0.1984612*** 0.1984853*** 0.1983364*** 0.1982436*** 0.1981224*** 0.1978182*** 0.1977608*** 0.1981543*** 0.1977221*** 0.1974126*** 0.19707316*** ROA_2023 -0.1705435*** -0.1701236*** -0.1702031*** -0.1703078*** -0.1705158*** -0.1705844*** -0.1710610*** -0.1711579*** -0.1706126*** -0.1712265*** -0.1716460*** -0.17209274*** LOW_30 -0.0002377 MEDIUM_30 -0.0002756 MEDIUMHIGH_30 -0.0002718 HIGH_30 -0.0026845* LOW_40 -0.0002293 MEDIUM_40 -0.0002169 MEDIUMHIGH_40 -0.0001725 HIGH_40 -0.0001646 LOW_50 -0.0002169 MEDIUM_50 -0.0001584 MEDIUMHIGH_50 -0.0001142 HIGH_50 -0.00007406 LN_EMISIONES_2022 -0.0026960* -0.0026810* -0.0026849* -0.0002506 -0.0026977* -0.0026970* -0.0027104* -0.0027095* -0.0027017* -0.0027117* -0.0027248* -0.00273084*
    R-squared 0.4457 0.4458 0.4458 0.4457 0.4457 0.4457 0.4456 0.4456 0.4457 0.4456 0.4456 0.4456

    Model Model 3. EMISSIONSREL 1 Model 3. EMISSIONSREL 2 Model 3. EMISSIONSREL 3 Model 3. EMISSIONSREL 4 Model 3. EMISSIONSREL 5 Model 3. EMISSIONSREL 6 Model 3. EMISSIONSREL 7 Model 3. EMISSIONSREL 8 Model 3. EMISSIONSREL 9 Model 3. EMISSIONSREL 10 Model 3. EMISSIONSREL 11 Model 3. EMISSIONSREL 12 intercept 0.1269910*** 0.1247362*** 0.1297759*** 0.1262328*** 0.1303181*** 0.1283068*** 0.1316277*** 0.1337368*** 0.1315412*** 0.1321394*** 0.1404204*** 0.1413043*** PD2022 0.5050857*** 0.5052116*** 0.5051322*** 0.5052150*** 0.5047766*** 0.5049183*** 0.5046450*** 0.5046948*** 0.5047595*** 0.5045969*** 0.5046710*** 0.5046256*** size_2023 -0.0054980** -0.0054822** -0.0054595** -0.0054793** -0.0054586** -0.0054596** -0.0054664** -0.0054360** -0.0054476** -0.0054390** -0.0054440** -0.0054351** BTM_2023 -0.0005158 -0.0005175 -0.0005108 -0.0005081 -0.0005177 -0.0005185 -0.0005237 -0.0005244 -0.0005205 -0.0005263 -0.0005288 -0.0005320
    LV_2023 0.1870193*** 0.1870104*** 0.1869547*** 0.1869940*** 0.1870622*** 0.1869567*** 0.1868944*** 0.1866786*** 0.1869872*** 0.1868847*** 0.1867495*** 0.1865755*** ROA_2023 -0.1861718*** -0.1858455*** -0.1860314*** -0.1857991*** -0.1858847*** -0.1857293*** -0.1860043*** -0.1862306*** -0.1858442*** -0.1862215*** -0.1861325*** -0.1864395*** LOW_30 -0.0010646* MEDIUM_30 -0.0010444* MEDIUMHIGH_30 -0.0011217** HIGH_30 -0.0010431* LOW_40 -0.0010937** MEDIUM_40 -0.0010496* MEDIUMHIGH_40 -0.0010738* HIGH_40 -0.0010882* LOW_50 -0.0011020** MEDIUM_50 -0.0010685* MEDIUMHIGH_50 -0.0011522** HIGH_50 -0.0011323*
    LN_EMISSIONSREL_2022 -0.0023691 -0.0023513 -0.0023066 -0.0023305 -0.0023325 -0.0023475 -0.0023498 -0.0023128 -0.0023318 -0.0023374 -0.0023063 -0.0023053
    R-squared 0.3885 0.3886 0.3888 0.3886 0.3887 0.3885 0.3885 0.3886 0.3887 0.3885 0.3886 0.3885

      Model Model 4. EMISSIONSREL 1 Model 4. EMISSIONSREL 2 Model 4. EMISSIONSREL 3 Model 4. EMISSIONSREL 4 Model 4. EMISSIONSREL 5 Model 4. EMISSIONSREL 6 Model 4. EMISSIONSREL 7 Model 4. EMISSIONSREL 8 Model 4. EMISSIONSREL 9 Model 4. EMISSIONSREL 10 Model 4. EMISSIONSREL 11 Model 4. EMISSIONSREL 12 intercept PD2022 0.4982708*** 0.4982774*** 0.4982263*** 0.4982801*** 0.4982179*** 0.4982571*** 0.49828289*** 0.4982899*** 0.4982270*** 0.49829957*** 0.49838208*** 0.49846427*** size_2023 -0.0004997 -0.0003760 -0.0003800 -0.0004377 -0.0005123 -0.0005437 -0.00070258 -0.0007183 -0.0005497 -0.00074557 -0.00092354 -0.00107871
    BTM_2023 -0.0005483 -0.0005456 -0.0005456 -0.0005457 -0.0005494 -0.0005498 -0.00055360 -0.0005541 -0.0005506 -0.00055456 -0.00055759 -0.00055971
    LV_2023 0.1985856*** 0.1987812*** 0.1987800*** 0.1986798*** 0.1985708*** 0.1984980*** 0.19823225*** 0.1981912*** 0.1984984*** 0.19815664*** 0.19785772*** 0.19759744*** ROA_2023 -0.1766911*** -0.1762419*** -0.1763428*** -0.1764446*** -0.1767279*** -0.1767927*** -0.17732197*** -0.1773967*** -0.1768481*** -0.17747140*** -0.17799542*** -0.17843322*** LOW_30 -0.0001515 MEDIUM_30 -0.0001949 MEDIUMHIGH_30 -0.0001882 HIGH_30 -0.0001694 LOW_40 -0.0001410 MEDIUM_40 -0.0001306 MEDIUMHIGH_40 -0.00007926 HIGH_40 -0.0000724 LOW_50 -0.0001274 MEDIUM_50 -0.00006514 MEDIUMHIGH_50 -0.00001281 HIGH_50 0.00002892 LN_EMISSIONSREL_2022 -0.0026240. -0.0025818. -0.0025882. -0.0025998. -0.0026297. -0.0026369. -0.00268372. -0.0026876. -0.0026419. -0.00269508. -0.00274259. -0.00278059.
    R-squared 0.4455 0.4456 0.4456 0.4456 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455 0.4455   Model Model 3. ENV_INV_INI 1 Model 3. ENV_INV_INI 2 Model 3. ENV_INV_INI 3 Model 3. ENV_INV_INI 4 Model 3. ENV_INV_INI 5 Model 3. ENV_INV_INI 6 Model 3. ENV_INV_INI 7 Model 3. ENV_INV_INI 8 Model 3. ENV_INV_INI 9 Model 3. ENV_INV_INI 10 Model 3. ENV_INV_INI 11 Model 3. ENV_INV_INI 12 intercept 0.09181814** 0.08955373** 0.09427877** 0.09187821** 0.09580378** 0.09398086** 0.09765361** 0.09934115** 0.09688006** 0.09834841** 0.10747123** 0.1083332** PD2022 0.52620556*** 0.52624538*** 0.52616975*** 0.52626514*** 0.52576702*** 0.52594113*** 0.52566443*** 0.52566269*** 0.52578115*** 0.52560058*** 0.52559636*** 0.5255627*** size_2023 -0.00321405* -0.00320779* -0.00318586. -0.00320487. -0.00317520. -0.00317908. -0.00316930. -0.00313889. -0.00316592. -0.00313878. -0.00313815. -0.0031261.
    BTM_2023 -0.00004537 -0.00004609 -0.00003865 -0.00003416 -0.00004748 -0.00004763 -0.00005396 -0.00005278 -0.00005062 -0.00005494 -0.00005784 -0.0000606
    LV_2023 0.17271723*** 0.17280254*** 0.17279489*** 0.17284058*** 0.17286369*** 0.17278426*** 0.17276424*** 0.17266728*** 0.17276698*** 0.17277580*** 0.17268178*** 0.1725444*** ROA_2023 -0.09077967*** -0.09062646*** -0.09064974*** -0.09052490*** -0.09069279*** -0.09067223*** -0.09071497*** -0.09079697*** -0.09067613*** -0.09084464*** -0.09082943*** -0.0908865*** LOW_30 -0.00118256** MEDIUM_30 -0.00116086** MEDIUMHIGH_30 -0.00122735** HIGH_30 -0.00117155** LOW_40 -0.00121719** MEDIUM_40 -0.00117556** MEDIUMHIGH_40 -0.00120688** HIGH_40 -0.00120919** LOW_50 -0.00122014** MEDIUM_50 -0.00120218** MEDIUMHIGH_50 -0.00129351** HIGH_50 -0.0012697** ENV_INV_INI_2022Y -0.01079555 -0.01073313 -0.01068283 -0.01063678 -0.01076058 -0.01080412 -0.01089093 -0.01086163 -0.01080057 -0.01090849 -0.01085645 -0.0109473
    R-squared 0.3839 0.3839 0.3841 0.384 0.3841 0.3839 0.3839 0.3839 0.384 0.3839 0.3841 0.3839

      Model Model 4. ENV_INV_INI 1 Model 4. ENV_INV_INI 2 Model 4. ENV_INV_INI 3 Model 4. ENV_INV_INI 4 Model 4. ENV_INV_INI 5 Model 4. ENV_INV_INI 6 Model 4. ENV_INV_INI 7 Model 4. ENV_INV_INI 8 Model 4. ENV_INV_INI 9 Model 4. ENV_INV_INI 10 Model 4. ENV_INV_INI 11 Model 4. ENV_INV_INI 12 intercept PD2022 0.52620556*** 0.52624538*** 0.52616975*** 0.52626514*** 0.52576702*** 0.52594113*** 0.52566443*** 0.52566269*** 0.52578115*** 0.52560058*** 0.52559636*** 0.5255627*** size_2023 -0.00321405* -0.00320779* -0.00318586. -0.00320487. -0.00317520. -0.00317908. -0.00316930. -0.00313889. -0.00316592. -0.00313878. -0.00313815. -0.0031261.
    BTM_2023 -0.00004537 -0.00004609 -0.00003865 -0.00003416 -0.00004748 -0.00004763 -0.00005396 -0.00005278 -0.00005062 -0.00005494 -0.00005784 -0.0000606
    LV_2023 0.17271723*** 0.17280254*** 0.17279489*** 0.17284058*** 0.17286369*** 0.17278426*** 0.17276424*** 0.17266728*** 0.17276698*** 0.17277580*** 0.17268178*** 0.1725444*** ROA_2023 -0.09077967*** -0.09062646*** -0.09064974*** -0.09052490*** -0.09069279*** -0.09067223*** -0.09071497*** -0.09079697*** -0.09067613*** -0.09084464*** -0.09082943*** -0.0908865*** LOW_30 -0.00118256** MEDIUM_30 -0.00116086** MEDIUMHIGH_30 -0.00122735** HIGH_30 -0.00117155** LOW_40 -0.00121719** MEDIUM_40 -0.00117556** MEDIUMHIGH_40 -0.00120688** HIGH_40 -0.00120919** LOW_50 -0.00122014** MEDIUM_50 -0.00120218** MEDIUMHIGH_50 -0.00129351** HIGH_50 -0.0012697** ENV_INV_INI_2022N 0.09181814** 0.08955373** 0.09427877** 0.09187821** 0.09580378** 0.09398086** 0.09765361** 0.09934115** 0.09688006** 0.09834841** 0.10747123** 0.1083332** ENV_INV_INI_2022Y 0.08102260* 0.07882060* 0.08359594* 0.08124143* 0.08504320* 0.08317673* 0.08676268* 0.08847952** 0.08607949* 0.08743992* 0.09661479** 0.0973859** R-squared 0.4465 0.4466 0.4467 0.4466 0.4467 0.4466 0.4465 0.4466 0.4467 0.4465 0.4467 0.4466

    Model Model 3. COMP_SUST 1 Model 3. COMP_SUST 2 Model 3. COMP_SUST 3 Model 3. COMP_SUST 4 Model 3. COMP_SUST 5 Model 3. COMP_SUST 6 Model 3. COMP_SUST 7 Model 3. COMP_SUST 8 Model 3. COMP_SUST 9 Model 3. COMP_SUST 10 Model 3. COMP_SUST 11 Model 3. COMP_SUST 12 intercept 0.1007621*** 0.09835028*** 0.10313436*** 0.10069735*** 0.10480646*** 0.1029617*** 0.10683185*** 0.10853926*** 0.10594653*** 0.10756161*** 0.11681148*** 0.1178206*** PD2022 0.5278887*** 0.52791760*** 0.52783041*** 0.52792332*** 0.52742695*** 0.5276140*** 0.52734313*** 0.52733384*** 0.52745122*** 0.52727732*** 0.52726256*** 0.5272406*** size_2023 -0.0040762* -0.00406212* -0.00403602* -0.00405027* -0.00403326* -0.0040398* -0.00404029* -0.00400345* -0.00403036* -0.00400810* -0.00400169* -0.0039957*
    BTM_2023 0.0000838 0.00008199 0.00008903 0.00009323 0.00008093 0.0000813 0.00007604 0.00007652 0.00007862 0.00007489 0.00007097 0.0000690
    LV_2023 0.1723753*** 0.17246645*** 0.17246024*** 0.17251504*** 0.17251936*** 0.1724407*** 0.17240873*** 0.17230889*** 0.17242089*** 0.17241537*** 0.17232514*** 0.1721716*** ROA_2023 -0.0912498*** -0.09108856*** -0.09111133*** -0.09097113*** -0.09117036*** -0.0911450*** -0.09120279*** -0.09128829*** -0.09115008*** -0.09134472*** -0.09132262*** -0.0913965*** LOW_30 -0.0012247** MEDIUM_30 -0.00120191** MEDIUMHIGH_30 -0.00126914** HIGH_30 -0.00121329** LOW_40 -0.00125914** MEDIUM_40 -0.0012165** MEDIUMHIGH_40 -0.00124870** HIGH_40 -0.00125093** LOW_50 -0.00126199** MEDIUM_50 -0.00124386** MEDIUMHIGH_50 -0.00133580** HIGH_50 -0.0013121** COMP_SUST_2022Y 0.0050404 0.00496997 0.00494940 0.00492037 0.00500833 0.0050187 0.00510544 0.00504067 0.00508592 0.00507363 0.00501398 0.0050355
    R-squared 0.3833 0.3834 0.3836 0.3835 0.3835 0.3834 0.3834 0.3834 0.3835 0.3833 0.3835 0.3834   Model Model 4. COMP_SUST 1 Model 4. COMP_SUST 2 Model 4. COMP_SUST 3 Model 4. COMP_SUST 4 Model 4. COMP_SUST 5 Model 4. COMP_SUST 6 Model 4. COMP_SUST 7 Model 4. COMP_SUST 8 Model 4. COMP_SUST 9 Model 4. COMP_SUST 10 Model 4. COMP_SUST 11 Model 4. COMP_SUST 12 intercept PD2022 0.5278887*** 0.52791760*** 0.52783041*** 0.52792332*** 0.52742695*** 0.5276140*** 0.52734313*** 0.52733384*** 0.52745122*** 0.52727732*** 0.52726256*** 0.5272406*** size_2023 -0.0040762* -0.00406212* -0.00403602* -0.00405027* -0.00403326* -0.0040398* -0.00404029* -0.00400345* -0.00403036* -0.00400810* -0.00400169* -0.0039957*
    BTM_2023 0.0000838 0.00008199 0.00008903 0.00009323 0.00008093 0.0000813 0.00007604 0.00007652 0.00007862 0.00007489 0.00007097 0.0000690
    LV_2023 0.1723753*** 0.17246645*** 0.17246024*** 0.17251504*** 0.17251936*** 0.1724407*** 0.17240873*** 0.17230889*** 0.17242089*** 0.17241537*** 0.17232514*** 0.1721716*** ROA_2023 -0.0912498*** -0.09108856*** -0.09111133*** -0.09097113*** -0.09117036*** -0.0911450*** -0.09120279*** -0.09128829*** -0.09115008*** -0.09134472*** -0.09132262*** -0.0913965*** LOW_30 -0.0012247** MEDIUM_30 -0.00120191** MEDIUMHIGH_30 -0.00126914** HIGH_30 -0.00121329** LOW_40 -0.00125914** MEDIUM_40 -0.0012165** MEDIUMHIGH_40 -0.00124870** HIGH_40 -0.00125093** LOW_50 -0.00126199** MEDIUM_50 -0.00124386** MEDIUMHIGH_50 -0.00133580** HIGH_50 -0.0013121** COMP_SUST_2022N 0.1007621*** 0.09835028*** 0.10313436*** 0.10069735*** 0.10480646*** 0.1029617*** 0.10683185*** 0.10853926*** 0.10594653*** 0.10756161*** 0.11681148*** 0.1178206*** COMP_SUST_2022Y 0.1058025*** 0.10332025*** 0.10808377*** 0.10561772*** 0.10981479*** 0.1079804*** 0.11193729*** 0.11357994*** 0.11103245*** 0.11263524*** 0.12182546*** 0.1228561*** R-squared 0.446 0.446 0.4462 0.4461 0.4461 0.446 0.446 0.446 0.4461 0.446 0.4462 0.446

  20. Assistant

    Sure, here is the interpretation and analysis written in English:


    Interpretation of Emissions and Environmental Actions

    Models with Intercept:

    1. 2030 Scenarios:

      • For the year 2030, companies under the LOW physical risk scenario have a significantly lower Probability of Default (PD) compared to those under the HIGH physical risk scenario. This suggests that in a low physical risk scenario, PD is greatly reduced, which is logical since lower risk scenarios result in lower probabilities of default.
    2. 2040 Scenarios:

      • In the year 2040, PD appears relatively stable across different physical risk scenarios. This consistency indicates that companies have adapted effectively to varying degrees of physical risks from previous years, maintaining a stable PD regardless of the physical risk level.
    3. 2050 Scenarios:

      • By the year 2050, the relationship reverses, with LOW physical risk scenarios exhibiting a higher PD than HIGH physical risk scenarios. This might seem counterintuitive economically, as higher risk scenarios should typically lead to higher PD. However, at such long-term horizons, this could reflect the possibility that companies in high-risk scenarios have developed more robust risk management practices over time, effectively reducing their PD, while companies in low-risk scenarios might have become complacent.

    General Insight:

    • Across all scenarios, it is evident that a higher composite score reduces the probability of default. This finding underscores the importance of enhancing a company’s risk management practices, indicating that as companies improve their risk scores or management strategies, their PD decreases significantly, leading to better financial outcomes and increased stability.

    Models without Intercept:

    In models without intercept, the coefficients for physical risk scenarios (LOW_30, MEDIUM_30, MEDIUMHIGH_30, and HIGH_30) become statistically insignificant. This indicates that physical risk scenarios do not significantly explain PD2023’s variability when the intercept is excluded, highlighting the importance of baseline factors. However, in these models, PD appears to decrease as the physical risk scenario increases.

    General Insight:

    • The significant drop in the explanatory power of physical risk variables in the absence of the intercept underscores the crucial role of baseline factors in explaining PD2023. The consistent insignificance of physical risk scenarios in models without intercept highlights the fundamental importance of including comprehensive baseline variables for precise analysis.

    Interpretation of Emissions and Environmental Actions

    Model Specifications:

    Using models from 3 and 4, the various emissions and environmental actions have been analyzed under different scenarios to assess their impact on PD2023.

    Results and Interpretation:

    1. Absolute Emissions (LN_EMISIONES_2022):

      • Across all Model 3 scenarios (e.g., EMISIONES 1 to 12) and Model 4 scenarios (e.g., EMISIONES 1 to 12), LN_EMISIONES_2022 consistently shows a negative coefficient. This indicates that higher absolute emissions are associated with a decrease in PD2023.
      • R-squared values range from 0.4129 to 0.4132 in Model 3 and from 0.4457 to 0.4458 in Model 4, suggesting that while emissions contribute to explaining the variability in PD, other factors also play significant roles.
    2. Relative Emissions (LN_EMISSIONSREL_2022):

      • Similar to absolute emissions, relative emissions (LN_EMISSIONSREL_2022) show negative coefficients across Model 3 scenarios (e.g., EMISSIONSREL 1 to 12) and Model 4 scenarios (e.g., EMISSIONSREL 1 to 12), indicating that higher relative emissions correlate with lower PD2023.
      • R-squared values remain consistent, indicating relative emissions' robust impact on explaining PD variability.
    3. Environmental Initiatives (ENV_INV_INI_2022):

      • Models incorporating environmental initiatives (e.g., ENV_INV_INI 1 to 12) demonstrate significant effects, with coefficients generally close to zero but negative, suggesting that companies investing in environmental initiatives may reduce their PD.
      • The R-squared values (around 0.3839 to 0.3841) indicate that these actions somewhat explain the variance in PD when considering other control variables.
    4. Compensation for Sustainability (COMP_SUST_2022):

      • Models including sustainability compensation (e.g., COMP_SUST 1 to 12) show positive and significant coefficients, suggesting that sustainability compensation initiatives are positively correlated with better PD outcomes.
      • R-squared values around 0.3833 to 0.3836 reflect consistent explanatory power, emphasizing the significance of sustainability compensation.

    General Insights:

    • Absolute and Relative Emissions: Both absolute and relative emissions display significant negative coefficients, indicating that higher emissions correlate with lower PD2023. This might reflect better environmental practices leading to greater operational efficiencies or better stakeholder relationships, thereby reducing financial risk.

    • Environmental Initiatives: Companies proactively investing in environmental initiatives see reductions in PD, underscoring the importance of such investments in risk management strategies.

    • Compensation for Sustainability: The positive relationship between sustainability compensation and PD suggests that aligning compensation with sustainability goals positively influences company performance, enhancing resilience and stability.

    Conclusion:

    Emissions and environmental actions significantly influence PD2023, revealing that proactive environmental responsibility can lead to differential financial outcomes. Companies that focus on reducing emissions, investing in environmental initiatives, and incorporating sustainability into compensation structures are likely to experience lower default probabilities, enhanced financial stability, and better long-term performance. This analysis underscores the importance of integrating environmental responsibility into corporate strategy, not just as a compliance measure but as a driver of financial resilience and success.