Wirte me a Matlab code to compute the efficient frontier for 3 assets. The assets are…
- User
Wirte me a Matlab code to compute the efficient frontier for 3 assets. The assets are perfectly correlated, have returns 2%, 6%, 7% and volatilities of 2%, 4%, and 7%. As output I want the frontier with the assets shown on the same graph, but also a chart with the portfolio composition along the frontier.
Assistant
I'll create a MATLAB code to compute and visualize the efficient frontier for your three assets. Note that since the assets are perfectly correlated, the efficient frontier will be linear rather than curved.
% Clear workspace and close figures clear all; close all; clc; % Asset parameters returns = [0.02; 0.06; 0.07]; % Expected returns volatilities = [0.02; 0.04; 0.07]; % Standard deviations n_assets = length(returns); % Correlation matrix (perfectly correlated) correlation = ones(n_assets); % Covariance matrix covariance = zeros(n_assets); for i = 1:n_assets for j = 1:n_assets covariance(i,j) = correlation(i,j) * volatilities(i) * volatilities(j); end end % Generate target returns for the efficient frontier n_points = 100; target_returns = linspace(min(returns), max(returns), n_points); % Initialize arrays for results frontier_volatility = zeros(n_points, 1); frontier_weights = zeros(n_points, n_assets); % Compute efficient frontier for i = 1:n_points % Minimize portfolio variance subject to target return % min w'*Cov*w subject to w'*r = target_return and sum(w) = 1 % Quadratic programming formulation H = 2 * covariance; f = zeros(n_assets, 1); % Equality constraints: Aeq*w = beq Aeq = [returns'; ones(1, n_assets)]; beq = [target_returns(i); 1]; % No short-selling constraint (optional - remove if short-selling allowed) lb = zeros(n_assets, 1); ub = ones(n_assets, 1); % Solve optimization options = optimoptions('quadprog', 'Display', 'off'); [w, fval] = quadprog(H, f, [], [], Aeq, beq, lb, ub, [], options); % Store results frontier_weights(i, :) = w'; frontier_volatility(i) = sqrt(w' * covariance * w); end % Figure 1: Efficient Frontier with Individual Assets figure('Position', [100, 100, 800, 600]); hold on; % Plot efficient frontier plot(frontier_volatility * 100, target_returns * 100, 'b-', 'LineWidth', 2); % Plot individual assets scatter(volatilities * 100, returns * 100, 100, 'filled'); text(volatilities(1)*100 + 0.2, returns(1)*100, 'Asset 1', 'FontSize', 12); text(volatilities(2)*100 + 0.2, returns(2)*100, 'Asset 2', 'FontSize', 12); text(volatilities(3)*100 + 0.2, returns(3)*100, 'Asset 3', 'FontSize', 12); % Labels and formatting xlabel('Volatility (%)', 'FontSize', 14); ylabel('Expected Return (%)', 'FontSize', 14); title('Efficient Frontier with Individual Assets', 'FontSize', 16); grid on; legend('Efficient Frontier', 'Individual Assets', 'Location', 'northwest'); hold off; % Figure 2: Portfolio Composition Along the Frontier figure('Position', [950, 100, 800, 600]); % Create area plot for portfolio weights area(target_returns * 100, frontier_weights * 100); % Labels and formatting xlabel('Expected Return (%)', 'FontSize', 14); ylabel('Portfolio Weight (%)', 'FontSize', 14); title('Portfolio Composition Along the Efficient Frontier', 'FontSize', 16); legend('Asset 1', 'Asset 2', 'Asset 3', 'Location', 'eastoutside'); grid on; ylim([0, 100]); % Display some key statistics fprintf('Asset Statistics:\n'); fprintf('================\n'); for i = 1:n_assets fprintf('Asset %d: Return = %.1f%%, Volatility = %.1f%%\n', ... i, returns(i)*100, volatilities(i)*100); end fprintf('\nNote: Since assets are perfectly correlated, the efficient frontier is linear.\n'); % Optional: Display portfolio weights at specific return levels fprintf('\nPortfolio Weights at Key Points:\n'); fprintf('================================\n'); key_returns = [0.02, 0.04, 0.06, 0.07]; for r = key_returns idx = find(abs(target_returns - r) < 0.0001, 1); if ~isempty(idx) fprintf('Return = %.1f%%: ', r*100); fprintf('w1 = %.1f%%, w2 = %.1f%%, w3 = %.1f%%\n', ... frontier_weights(idx,:)*100); end endThis code will generate two figures:
-
Efficient Frontier Plot: Shows the efficient frontier as a line with the three individual assets plotted as points. Since the assets are perfectly correlated, the frontier will be a straight line connecting the assets.
-
Portfolio Composition Chart: An area chart showing how the portfolio weights change as you move along the efficient frontier from the lowest return to the highest return.
Key points about the code:
- Uses quadratic programming to solve the mean-variance optimization problem
- Includes a no short-selling constraint (you can remove the
lbandubconstraints if you want to allow short positions) - Since the assets are perfectly correlated, the efficient frontier will be linear, consisting of combinations of only two assets at any point
- The portfolio composition will show a transition from Asset 1 (lowest risk/return) to Asset 3 (highest risk/return)
The perfect correlation means that diversification benefits are eliminated, and the optimal portfolios will typically consist of combinations of adjacent assets on the risk-return spectrum.
-
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.