Skip to content
All library documents

Interpreting Call Delta Hedging in a Binomial Model

Article Quant Q&A · Author: Snowflake

Summary

The document discusses why the binomial call option delta implies that a portfolio holding delta shares and short one call has the same value in the up and down states. Delta is the change in option value divided by the change in stock price, so the equality follows algebraically from its definition. In a one-period binomial model, that equality describes a locally hedged position whose terminal value is insensitive to which of the two modeled stock moves occurs.

The answers connect the relation to delta neutrality and replication, while emphasizing limits. It applies at each step when delta is calculated for that step's possible outcomes; it does not mean delta stays constant as prices move. Since an option is nonlinear, a hedge based on a fixed delta is only locally accurate and may need rebalancing after the underlying changes. The discussion is conceptual and provides no market data or empirical test.

Key ideas

  • In a binomial model, delta is the slope between the option values in the up and down states.
  • The delta definition makes the stock-and-short-call portfolio value equal across those two states.
  • Delta can serve as the stock weight in a replicating portfolio.
  • A fixed delta hedge is only locally accurate because option values are nonlinear.

Tags

Full text
# Interpretation of equation derived from the delta of a call European call option


# Interpretation of equation derived from the delta of a call European call option












I have started reading an introductory book called: A Course in Derivative Securities by Kerry Back. On page 12 they mention the following:

The delta of the call option is $\delta = (C_{u} - C_{d}) / (S_{u} - S_{d})$ and then they rewrite this to $\delta S_{u} - C_{u} = \delta S_{d} - C_{d}$, where $S_{u}$ is the stock price in the "up state", $S_{d}$ for the "down state" and $C_{u} = max(0, S_{u} - K)$, $K$ is the exercise price.

Now I am wondering, besides from the math, why is it, intuitively, that $\delta S_{u} - C_{u} = \delta S_{d} - C_{d}$ on day 0. Does this also hold on any other day? If so, could someone intuitively tell me why (I get the derivation though, but I lack the deeper understanding of why).

Thanks a lot!

## Answer by emcor (score 1, accepted)

https://quant.stackexchange.com/a/14041

Call Delta is generally defined as $$\Delta_C=\partial_S C=\frac{dC}{dS}=\frac{C_u-C_d}{S_u-S_d}$$, so it is the derivative or tangential change in $C$ from change in $S$, discretized in the Binomial Model.

As we know, this derivative goes symmetrically both ways, when $C_u$ goes up or $C_d$ down, so one can in general rewrite this equation:

$$\Delta S_u-C_u=\Delta S_d-C_d$$

As explained, this holds on every day just by the definition of $\Delta$, but it has no direct intuitive interpretation yet as such. You may say that for no-arbitrage, a symmetric structure is in general "better", but the result here is just a consequence of discretizing the derivative in $\Delta$. Later on, one may find that Delta also happens to be the stock-weight for a replicating portfolio.

## Answer by Matt B. (score 1)

https://quant.stackexchange.com/a/14044

Intuitively, holding $\delta$ stocks in your portfolio is going to make you money if the stock goes up (but you're going to lose on the option you've sold), and lose you money if the stock goes down (but you make on the option which becomes worthless).

The equality is the basis for the concept of $\delta$-neutrality, ie whatever happens, your portfolio value is unaffected, and is equal to today's value (discounted by the risk free rate)

## Answer by Kumar (score 0)

https://quant.stackexchange.com/a/14047

Its not the day that matters. Delta will remain the same on subsequent days if the price of the option and its underlying remain unchanged since the time the delta was calculated.

Delta is a linear approximation and measures only the slope, but options are non-linear instruments, they have a convexity. Delta holds only for small changes in underlying so any large change in underlying means that delta needs to be revalued.

Also, $\delta S_u - C_u = \delta S_d - C_d$ holds by definition of delta hedging and how delta is evaluated.

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.