The Mean Value Theorem states that for f continuous on [a,b] and differentiable on (a,b) there is an interior point c with f(b) − f(a) = f'(c)(b − a). Suppose you don't know that derivative exactly, but you do know it never gets bigger than some number M in size: |f'(x)| ≤ M for all x in the interval. Take absolute values and the unknown slope is replaced by that ceiling: |f(b) − f(a)| = |f'(c)||b − a| ≤ M|b − a|. So a bound on the rate of change becomes a bound on how far the function's value can travel — the standard way to turn a derivative estimate into a numerical error bound.
Rolle's Theorem: The Flat-Point Guarantee and Where It Pays Off
Error Bounds, Sensitivity, and Applied Uses
From a Slope Bound to a Value Bound
1 / 5
The Mean Value Theorem gave us one equation: the change in the function equals the derivative at some interior point, times the width of the interval. Now suppose we don't know that derivative exactly, but we do know it can never be larger than some number M in size.
0:00 / 0:00