Skip to content
Derivatives Pricing and Computational Finance / Foundations of Computational Option Pricing

Derivation: Transforming Black-Scholes to the Heat Equation

The Black–Scholes PDE has variable coefficients (SS, S2S^2) and runs backwards in time. The classical heat equation yτ=yxxy_\tau = y_{xx} has constant coefficients and runs forwards. This subnote walks through the substitutions that map one into the other, explains why each substitution is the natural one, and records the transformed payoffs and boundary data used by the finite-difference schemes of finite difference methods. The reward for the algebra: every piece of numerical analysis (stability, convergence, scheme design) can be done once, on the simplest parabolic PDE there is, then transported back to finance.

In this lesson

  1. Step 0 — Starting point
  2. Step 1 — Why S=KexS = Ke^x: kill the variable coefficients
  3. Step 2 — Why τ=12σ2(T−t)\tau = \tfrac12\sigma^2 (T-t): forward time, natural units
  4. Step 3 — Why the exponential rescaling: absorb drift and discounting
  5. Step 4 — Transformed payoffs
  6. Step 5 — Transformed boundary conditions
  7. What the transformation buys

The full lesson, with worked examples and exercise solutions, is for members.

Read the full lesson

New to SelfMadeQuant.com? See membership options or browse the Derivatives Pricing and Computational Finance syllabus.