I'm reading Option Pricing: A Simplified Approach and have a question. Assume the binomial tree model for the stock. So

  • $n$ discrete time periods
  • $S$ is stock
  • $C$ is call
  • $K$ is strike
  • $u$ is upward move
  • $d$ is downward move
  • $r$ is total return $1+R$ where $R$ is the interest rate
  • no arb: $d < r < u$
  • $q$ is probability of upward move
  • $p = (r-d) / (u-d)$ is risk-neutral probability of upward move

Fine so far. But then Cox defines $p^{\prime}$ as $(u/r) p$. And I don't understand what this parameter represents. Concretely, the binomial options pricing formula for a call $C$ is

$$ C = S \Phi(a; n, p^{\prime}) - K r^{-n} \Phi(a; n, p) $$

where $a$ is the min moves for the call to be in-the-money and $\Phi$ is the complementary binomial distribution function, i.e. 1 minus the CDF. I can't make sense of $p^{\prime}$ though. Why does this adjustment factor fall out of the model? What does it represent?

  • $\begingroup$ Intuitively when the option expires ITM (which happens with probability $p$) you will not receive stock worth S (the current stock price) but worth an amount $S' > S$ and you will be in a situation where the stock has appreciated more than expected by the factor $u/r$. $\endgroup$
    – nbbo2
    Commented May 8, 2023 at 6:05
  • $\begingroup$ You could handle the situation by using the enhanced price $S′$ in the call price formula, instead Cox uses $S$ and handles it by using an "enhanced probability" $p′$. What matters is the product of the two, so both methods give the same result. $\endgroup$
    – nbbo2
    Commented May 8, 2023 at 12:03
  • $\begingroup$ This makes sense, thanks. What made it click for me was computing the expected value after one period, so $E[S] = (puS + qdS) / r$. $\endgroup$
    – jds
    Commented May 9, 2023 at 11:06


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.