Skip to content
All library documents

Expanding the Squared Spot-Price Derivative in Black–Scholes

Article Quant Q&A · Author: LeptoSq

Summary

This note explains an operator identity used to rewrite the Black–Scholes equation before transforming it into the one-dimensional heat equation. The key is to interpret the squared operator as applying S times differentiation with respect to S twice, in sequence. Applying the product rule to that composition produces both a first-derivative term and a second-derivative term.

The resulting identity shows why the rewritten equation includes a compensating negative first-derivative term: it offsets the extra term introduced when expanding the squared operator. The answer also points out that differentiation does not commute with multiplication by S, which is the source of the original confusion. The discussion establishes this algebraic step, but does not carry out the subsequent change of variables or derive the heat equation itself.

Key ideas

  • The squared operator means applying S times the spot-price derivative twice in sequence.
  • The product rule gives a first-derivative term in addition to the second-derivative term.
  • Multiplication by S and differentiation with respect to S do not commute.
  • A compensating first-derivative term makes the rewritten Black–Scholes equation equivalent to its original form.

Tags

Full text
# Equivalent form of Black-Scholes Equation (to transform to heat equation)


# Equivalent form of Black-Scholes Equation (to transform to heat equation)












I am trying to understand the transformation of the Black-Scholes equation to the one-dimensional heat equation from Joshi, M. (2011). The Concepts and practice of mathematical finance. 2nd ed. Cambridge, U.K.: Cambridge University Press, pp.119. . The author stated that the Black-Scholes equation:

$ \frac{\partial C}{\partial t}(S,t) +rS \frac{\partial C}{\partial S}(S,t) + \frac{1}{2}\sigma^2 S^2 \frac{\partial^2 C}{\partial S^2}(S,t) -rC=0 $

can be rewritten as:

$ \frac{\partial C}{\partial t}(S,t) +(r-\frac{1}{2}\sigma^2) S \frac{\partial C}{\partial S}(S,t) + \frac{1}{2}\sigma^2 (S \frac{\partial}{\partial S})^2 C -rC=0 $,

however I can't see how these two equations are equivalent, in particular, I don't understand how the additional $-\frac{1}{2} \sigma^2 S \frac{\partial C}{\partial S} (S,t)$ came about. I have tried expanding the squared differential operator as follows:

\begin{align*} & \frac{1}{2}\sigma^2(S \frac{\partial}{\partial S})^2 C(S,t) \\ &= \frac{1}{2} \sigma^2 \frac{\partial}{\partial S} [\frac{\partial}{\partial S} (S^2 C(S,t))] \\ &= \frac{1}{2}\sigma^2 \frac{\partial}{\partial S} [2S\cdot C(S,t) + S^2 \frac{\partial C(S,t)}{\partial S}] \\ &= \frac{1}{2}\sigma^2 [2C(S,t) + 2S \frac{\partial C(S,t)}{\partial S} + 2S \frac{\partial C(S,t)}{\partial S} + S^2 \frac{\partial^2 C(S,t)}{\partial S^2} \\ &= \frac{1}{2} \sigma^2 [2C(S,t) + 4S \frac{\partial C(S,t)}{\partial S}] + \frac{1}{2} \sigma^2 S^2 \frac{\partial ^2 C(S,t)}{\partial S^2} \end{align*}

Is there something that I missed out? Can anyone help me out?

Thanks.

## Answer by VommaNeutral (score 9, accepted)

https://quant.stackexchange.com/a/31840

I think you've cottoned on to the main question - that is,to see why $\frac{1}{2} \sigma^2 S^2 \frac{\partial^2 C}{ \partial S^2 }$ is the same thing as $-\frac{1}{2} \sigma^2 S \frac{\partial C}{\partial S} + \frac{1}{2} \sigma^2 (S \frac{\partial}{\partial S})^2 C $. I think your confusion comes from dealing with the squared operator.

Writing down what the expression means, we have

$$ (S \frac{\partial}{\partial S})^2 C= (S \frac{\partial}{\partial S}) (S \frac{\partial}{\partial S}) C $$

That is, apply $S \frac{\partial}{ \partial S} $ to $C$ first, then apply $S \frac{\partial}{\partial S}$ to the result. Note that we can't just switch the order of these symbols - this is not the same thing as $\frac{\partial}{\partial S} \frac{\partial}{\partial S} S^2 $. (The derivative operator rarely commutes with anything - try evaluating this on a hypothetical option with constant value $C=1$) Expanding out the expression should give

$$ (S \frac{\partial}{\partial S})^2 C = S \frac{\partial}{\partial S} (S \frac{\partial C}{\partial S}) $$

and applying the product rule gives that this is equal to

$$ S (\frac{\partial S}{\partial S} \frac{\partial C}{\partial S} + S \frac{\partial ^2 C}{\partial S ^2}) $$

(Incidentally, we can see that it's exactly the product rule that makes the derivative fail to commute with multiplication by $S$). So

$$ (S \frac{\partial}{\partial S})^2 C= S \frac{\partial C}{\partial S} + S^2 \frac{\partial ^2 C}{\partial S ^2} $$

Multiply this by $\frac{1}{2} \sigma^2$ and rearrange a little and we should have our result.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.