the Black-Scholes equation
/ blak SHOHLZ /
Suppose you want to buy the right (but not the obligation) to purchase a stock at a fixed price three months from now — a 'call option'. What is that right worth today, given that the stock price wanders unpredictably? The Black-Scholes equation answers this. It was the breakthrough that turned financial derivatives into a rigorous, mathematical discipline, and it is the heat equation wearing a pinstripe suit.
Let V(S, t) be the value of the option as a function of the underlying stock price S and time t. Under the model's assumptions, V satisfies V_t + (1/2) sigma^2 S^2 V_SS + r S V_S - r V = 0, where sigma is the stock's volatility (how jittery the price is) and r the risk-free interest rate. The crucial term is (1/2) sigma^2 S^2 V_SS, a diffusion term: uncertainty in the stock price spreads the option value out exactly as heat spreads. It is a backward parabolic equation — you know the value at expiry (the payoff, a known function of S) and solve backward in time to find today's value. A change of variables (take logs of S and reverse time) turns it literally into the standard heat equation u_tau = u_xx, which is why a clean closed-form price exists.
There is a deeper bridge here: the Feynman-Kac formula says the solution of such a parabolic PDE can be written as an expected value over random (Brownian) paths of the underlying. So pricing an option = averaging its discounted payoff over all the random ways the stock could wander, which is exactly how the diffusion-PDE and probability pictures meet. The honest caveat is large: the model assumes constant volatility, frictionless trading, and log-normal prices — assumptions markets routinely violate (the 'volatility smile' is the visible scar), so Black-Scholes is a brilliant baseline, not the truth.
With the substitution x = log S and reversed, rescaled time, the Black-Scholes equation becomes exactly u_tau = u_xx, the heat equation. The known closed-form 'Black-Scholes formula' for a European call is then nothing but the heat kernel (a Gaussian) convolved with the payoff — the same mathematics that smooths a temperature spike smooths the option's terminal payoff backward in time.
A backward heat equation; Feynman-Kac links it to averaging over diffusion paths.
Solving it backward in time is well-posed (terminal payoff given), unlike the genuinely ill-posed backward heat equation with initial data — direction matters. And its idealized assumptions (constant volatility, no jumps) are systematically wrong, which is why real desks bolt on corrections rather than trust the bare formula.