Why a One-Step Binomial Call Is Priced by Replication
Summary
The document examines a one-day call option on a stock priced at 100, with possible next-day prices of 110 or 90, and zero interest rates. It shows how to replicate the call payoff by combining the option sale proceeds with a position of one-half share. Matching the portfolio’s outcomes in both states gives a call price of 5. This price follows from no-arbitrage replication and does not depend on the stated 99% and 1% probabilities, provided the stock and option prices permit the hedge described.
The text contrasts that result with valuing the payoff using the stated probabilities, which gives 9.90, and includes a response that argues for this probability-weighted value. Another response stresses that the stock’s expected future price and its current market price raise a separate question involving investor risk tolerance and correlations with other assets. The discussion is brief and does not resolve all assumptions behind the disagreement; its central lesson is that physical probabilities alone do not override a valid replicating hedge’s no-arbitrage price.
Key ideas
- A call paying 10 or zero can be replicated with one-half share in the stated two-state setup.
- The replication equations imply a call value of 5 under the stated stock price and zero interest rate.
- The 99% probability of an upward move does not change the price implied by exact replication.
- The responses disagree with a probability-weighted valuation of 9.90, highlighting the distinction between expected payoff and arbitrage pricing.
- The stock’s own market price depends on broader investor preferences and portfolio relationships.
Tags
Full text
# Basic binomial option pricing example
# Basic binomial option pricing example
A security is currently trading at 100, and with 99% probability it will be at 110 tomorrow, and with 1% probability at 90. What is the value of an ATM call option today expiring tomorrow? Assume nil interest rates.
If you sell the option at $C$ and immediately hedge with $n$ shares of the underlying, you'll be left with $C - 100n$. Tomorrow your portfolio will be worth $C - 100n + 110n = C + 10n$ with 99% chance and $C - 100n + 90n = C-10n$ with 1% chance. In order for you to not lose or make money on selling the option, in the first case your portfolio must be worth 10, and in the second case 0. In other words, $C$ and $n$ are uniquely determined by
\begin{cases} C + 10n = 10 \\ C - 10n = 0. \end{cases} Solving yields $C = 5$ and $n = 1/2$.
On the other hand, one sees that there is a 99% chance the option has value 10 dollars tomorrow and a 1% chance it has no value, yielding $C = (10)(.99) = 9.90.$
Which is the correct valuation? Is the stock mispriced if we know there is a 99% chance of an upward movement?
## Answer by dm63 (score 1)
https://quant.stackexchange.com/a/35729
The correct valuation is C=5, assuming the stock is trading at 100. Any other option price allows an immediate arbitrage opportunity between the option and the stock.
Whether the stock should be trading at 100 is a separate question. It depends on the risk tolerance of investors, and the correlation of this stock with other investments, for example.
## Answer by will (score 0)
https://quant.stackexchange.com/a/35731
Your example is a bit odd;
tomorrow, you expect a stock to be worth either $110$, with $99\%$ probability, or $90$ with $1\%$ probability. So your expectation of the value tomorrow is $0.99\cdot110 + 0.01\cdot 90 = 109.8$ - this means that the intrinsic value of your option is $\max(0, 109.8-100) = 9.8$. If the option is worth less than this, then there is an arbitrage, as it has negative time value.
The value of the option is
$$\begin{eqnarray} \mathrm{Call} &=& \int_0^\infty \phi(S_T) \cdot \max(S_t-K,0)\ \mathrm{d}S_T\\ &=&\phi(110) \cdot (110-100)^+ + \phi(90)\cdot(90-100)^+ \\ &=& \phi(110) \cdot 10 = 99\% \cdot 10\\ &=& 9.9 \end{eqnarray}$$
This is an odd / contrived example though, since the spot price is so far from the forward, and we have said there are no dividends and interest rates are zero. So you're only going to be able to explain this with some fairly contrived scenarios (i.e. maybe the company is selling loads of stock to raise money and driving the price down, such that it's not at an equillibrium value, and the market knows this will stop tomorrow? Who knows.)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.