The interactive film is in the book's Companion

Play it, pause it, turn it and drag it with Mathematics for AI and Machine Learning.

Get the book

Itô's Lemma: The Extra Second-Order Drift

Chapter 19 · Mathematics for AI and Machine Learning · Advanced · 2:01

An interactive live film: f(W_t) - f(0) vs the first-order stochastic integral, with the discrepancy perfectly tracked by the accumulated curvature drift (1/2) f''(W_s) dt.

Narration

In ordinary calculus, the differential of f of x is f prime of x times d x. But when x is a Brownian motion, its quadratic variation is non zero, and ordinary calculus fails.

Ito's lemma expands f to second order. Because d W squared equals d t, the second derivative term does not vanish. It adds a deterministic drift of half f double prime d t.

The rest of the narration is in the book's Companion.

More from Chapter 19

Interactive1:37Brownian Motion: Rough Paths, a Spreading Bell, and dW Squared Equals dtChapter 19Preview1:17The Ornstein-Uhlenbeck Process: Brownian Motion on a SpringChapter 19Preview1:29Conditioning a Diffusion on Its EndpointChapter 19Preview1:46Itô vs Stratonovich: Left Endpoint or MidpointChapter 19Preview1:56Euler–Maruyama: SDEs as Limits of AlgorithmsChapter 19