Prove the mean value theorem for derivatives.
- User
Prove the mean value theorem for derivatives.
- Assistant
Crafting explanations for MVT and EVT
I’m noticing that many textbooks introduce the Mean Value Theorem (MVT) proof using the Extreme Value Theorem (EVT), which relies on the completeness of the real numbers. I think I'll focus on crafting a moderate-level explanation, providing necessary arguments without being overly verbose. Also, I need to remember to avoid generic phrases like "Great question" at the start—better to dive right into the topic. Alright, let's get to writing! Mean Value Theorem (for derivatives)
Let (f:[a,b]\to\mathbb{R}) be continuous on ([a,b]) and differentiable on ((a,b)).
Then there exists some (c\in(a,b)) such that [ f'(c) ;=; \frac{f(b)-f(a)}{b-a}. ]A standard proof proceeds via two steps:
- Fermat’s Theorem (derivative at an interior extremum is zero),
- Rolle’s Theorem, and then
- Derive the Mean Value Theorem from Rolle’s Theorem.
I’ll assume the Extreme Value Theorem: a continuous function on a closed interval ([a,b]) attains its maximum and minimum.
Step 1: Fermat’s Theorem
Theorem (Fermat).
Let (f) be differentiable at (c\in(a,b)). If (f) has a local maximum or minimum at (c), then [ f'(c) = 0. ]Proof.
Assume (c) is a local maximum. (The minimum case is analogous.)Then there exists (\delta>0) such that for all (x) with (|x-c|<\delta), [ f(x) \le f(c). ]
For (0<h<\delta), we have (c+h\in(a,b)) and [ \frac{f(c+h)-f(c)}{h} \le 0 ] since (f(c+h)\le f(c)) and (h>0).
For (-\delta<h<0) (so (h<0)), we have (c+h\in(a,b)) and [ \frac{f(c+h)-f(c)}{h} \ge 0 ] because (f(c+h)\le f(c)) and (h<0) flips the inequality.
By definition of the derivative, both one-sided difference quotients must converge to (f'(c)): [ \lim_{h\to 0^+} \frac{f(c+h)-f(c)}{h} = f'(c), \quad \lim_{h\to 0^-} \frac{f(c+h)-f(c)}{h} = f'(c). ]
But from the inequalities, [ \lim_{h\to 0^+} \frac{f(c+h)-f(c)}{h} \le 0,\quad \lim_{h\to 0^-} \frac{f(c+h)-f(c)}{h} \ge 0. ]
Since these two limits are equal (both equal (f'(c))), we must have [ f'(c) \le 0 \quad\text{and}\quad f'(c) \ge 0, ] so (f'(c) = 0).
This proves Fermat’s theorem.
Step 2: Rolle’s Theorem
Rolle’s Theorem.
Let (f:[a,b]\to\mathbb{R}) be continuous on ([a,b]), differentiable on ((a,b)), and satisfy (f(a) = f(b)).
Then there exists some (c\in(a,b)) such that [ f'(c) = 0. ]Proof.
-
By the Extreme Value Theorem, since (f) is continuous on the compact interval ([a,b]), it attains its maximum and minimum on ([a,b]). That is, there exist points (x_{\max}, x_{\min} \in [a,b]) such that [ f(x_{\max}) = \max_{x\in[a,b]} f(x),\quad f(x_{\min}) = \min_{x\in[a,b]} f(x). ]
-
If (f) is constant on ([a,b]), then (f'(x)=0) for all (x\in(a,b)), and the theorem holds trivially.
-
Otherwise, (f) is not constant, so either its maximum or its minimum is different from (f(a)=f(b)).
- If (f(x_{\max}) > f(a)), then (x_{\max}) cannot be (a) or (b) (since (f(a)=f(b))), so (x_{\max}\in(a,b)). Then (x_{\max}) is a local maximum in the interior, and by Fermat’s theorem, (f'(x_{\max})=0).
- If instead (f(x_{\min}) < f(a)), analogous reasoning shows (x_{\min} \in (a,b)), giving (f'(x_{\min})=0).
In either case, there is some (c\in(a,b)) such that (f'(c)=0). This is Rolle’s Theorem.
Step 3: Proof of the Mean Value Theorem
Mean Value Theorem.
Let (f:[a,b]\to\mathbb{R}) be continuous on ([a,b]) and differentiable on ((a,b)).
Then there is a point (c\in(a,b)) such that [ f'(c) = \frac{f(b)-f(a)}{b-a}. ]Proof.
Define a new function (g:[a,b]\to\mathbb{R}) by subtracting off the straight line between ((a,f(a))) and ((b,f(b))): [ g(x) := f(x) - \ell(x), ] where (\ell) is the linear function [ \ell(x) = f(a) + \frac{f(b)-f(a)}{b-a}(x-a). ]
Observe:
-
Continuity and differentiability of (g):
Since (f) is continuous on ([a,b]) and differentiable on ((a,b)), and (\ell) is a polynomial (hence continuous and differentiable everywhere), (g) is continuous on ([a,b]) and differentiable on ((a,b)). -
Equal values at the endpoints:
Compute [ g(a) = f(a) - \ell(a) = f(a) - \Bigl(f(a) + \frac{f(b)-f(a)}{b-a}(a-a)\Bigr) = f(a)-f(a) = 0, ] and [ g(b) = f(b) - \ell(b) = f(b) - \Bigl(f(a) + \frac{f(b)-f(a)}{b-a}(b-a)\Bigr) = f(b) - \bigl(f(a) + (f(b)-f(a))\bigr) = 0. ] So (g(a) = g(b)). -
Apply Rolle’s Theorem:
By Rolle’s Theorem (applied to (g)), there exists (c\in(a,b)) such that [ g'(c) = 0. ] -
Compute (g'(c)): Since (g(x) = f(x) - \ell(x)), [ g'(x) = f'(x) - \ell'(x). ] But (\ell'(x)) is just the constant slope of the line: [ \ell'(x) = \frac{f(b)-f(a)}{b-a}. ] Therefore, [ g'(c) = f'(c) - \frac{f(b)-f(a)}{b-a} = 0, ] which gives [ f'(c) = \frac{f(b)-f(a)}{b-a}. ]
This completes the proof of the Mean Value Theorem for derivatives.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.