Skip to content
All library documents

Delta Hedging and the No-Arbitrage Basis of Black–Scholes

Article Quant Q&A · Author: zzzbbx

Summary

The document explains why the Black–Scholes derivation forms a portfolio by shorting a derivative and holding a delta-sized position in the underlying stock. Delta is the derivative’s sensitivity to the stock price; choosing the stock position to offset that sensitivity makes the portfolio locally hedged. Under the model’s assumptions, no-arbitrage reasoning then requires the hedged portfolio to earn the risk-free rate, which helps derive the pricing equation.

The discussion clarifies that this portfolio is not simply equal in value to the option premium, and that the setup does not depend on whether the derivative is a call or a put. It also sketches a replication approach using cash and shares, with Itô’s formula linking portfolio changes to the Black–Scholes partial differential equation. These explanations rely on differentiability and the model’s assumptions; the text does not enumerate all assumptions or address market frictions. It notes that similar hedging arguments can extend to payoffs depending on multiple underlying assets by adding positions in those assets.

Key ideas

  • A delta hedge combines a short derivative position with an offsetting position in the underlying stock.
  • The hedged portfolio is used in the derivation because no-arbitrage assumptions constrain its return to the risk-free rate.
  • The portfolio’s value is not the same thing as the derivative premium.
  • The derivation applies to differentiable derivative prices and does not require the payoff to be a call.
  • A replication argument can be extended to claims with multiple underlying assets by including additional hedge positions.

Tags

Full text
# Basic question about Black Scholes derivation


# Basic question about Black Scholes derivation












In the derivation of the Black Scholes equation, the value of the portfolio at time $t$ is given by $$P_t = -D_t + \frac{{\partial D_t}}{{\partial S_t}}S_t $$ where $P_t$ is the value of the portfolio at time $t$, $D_t$ the price of the derivative and $S_t$ the price of the stock. I guess $D_t$ means the price of the premium, i.e., the right to by a call/put option.

1) Why is that the value of the portfolio is equal to the value of the premium? 2) Does this equation assume that the underlying derivative is a call option?

## Answer by jlowin (score 2)

https://quant.stackexchange.com/a/4005

1) Instead of asking why the portfolio is equal to the premium, ask why create it at all? I say that because it actually isn't equal to the premium, since that would just be D, and also because the actual value isn't as important as what the portfolio represents.

We form this particular portfolio because the laws of no-arbitrage guarantee it has a certain rate of return (within a set of necessary assumptions). That gives us a constant point which we can exploit to solve the equation; without this observation, the equations aren't grounded.

Specifically, the portfolio consists of a short derivative paired with an amount of stock such that the change in the derivative price due to the stock moving is exactly offset by these shares. That amount, represented here by the $\frac{{\partial D_t}}{{\partial S_t}}$ term, is called "delta". Because of this pairing, the resulting portfolio is not exposed to risk (i.e. it has been hedged) and therefore we know it must return the risk-free rate. In a later step of the derivation, we will differentiate the portfolio value with respect to time and use this information to figure out what that is (unsurprisingly, it is simply a proportionate piece of the risk-free rate).

So in sum: the portfolio is set up in such a way that we know how it behaves, a fact which we later use to anchor the differential equation.

2) This equation is valid for any derivative that is differentiable twice with respect to S (the stock price) and once with respect to t (time). We have made no assumptions yet about put or call. As you will see, delta for puts is between -1 and 0, so if the derivative were a put the stock component of this portfolio would be negative.

Hull's book really does a great job of explaining this, I suggest you pick up a copy if you haven't already.

## Answer by vanguard2k (score 0)

https://quant.stackexchange.com/a/4003

You are refering to a portfolio consisting of a call option and $\frac{\partial D_t}{\partial S_t}$ (the so-called Delta) "shares" of the "stock" $S_t$. This strategy is used to "replicate a bond" in the derivation. Since $P_t$ should replicate a bond you will have to use this in your further derivation (an thus get an equation). Just continue to read the derivation and it will all come out.

## Answer by vanna (score 0)

https://quant.stackexchange.com/a/4017

Say you have a portfolio with $\alpha$ dollars in cash and $\beta$ stocks at time $t=0$. The value of your portfolio at time $t$ is

$$P_t = \alpha e^{rt} + \beta S_t \tag{1} $$

Black-Scholes approach solve the following problem

> Can we find $(\alpha,\beta)$ such that the portfolio is self-financing and has final value equal to the payoff of the option ? If yes then -no arbitrage argument- the fair price of the option is the initial value of the portfolio.

The self-financing equation is (in continuous time) : $$ dP_t = \alpha r e^{rt} dt + \beta dS_t \tag{2}$$

The value of $\alpha$ and $\beta$ are found using a purely mathematical derivation. We assume that $P_t=P(t,S_t)$ and use Itô formula : $$ dP_t = \frac{\partial{P_t}}{{\partial t}} dt + \frac{\partial P_t}{\partial S} dS_t + \frac{1}{2} \frac{\partial^2 P_t}{{\partial S^2}} d\langle S,S \rangle_t \tag{3}$$

Taking $\beta=\frac{\partial P_t}{\partial S}$ in $(2)$ and equalizing $(2)$ and $(3)$ yields the Black-Scholes PDE. The value of $P_t$ is the solution of this PDE, and the delta is the derivative of this solution w.r.t. its second argument, $S$.

This approach does not use the fact that the option is a call option. You actually only need a payoff of form $f(S_T)$. If the payoff involves two underlying e.g. you can use similar approaches with $\alpha, \beta$ and $\gamma$ for the second underlying.

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.