Pricing a Down-and-In Cash-or-Nothing Option with Brownian Motion
Summary
The document considers a European cash-or-nothing option that pays when the terminal underlying value exceeds a strike, provided the process has first crossed a lower barrier. The underlying is modeled as driftless Brownian motion with constant volatility. The question proposes multiplying the ordinary binary option probability by the probability of hitting the barrier, using the reflection principle to calculate the latter.
The accepted response identifies the central flaw: terminal moneyness and barrier crossing are dependent events, so their probabilities cannot simply be multiplied. Instead, the desired value is the discounted risk-neutral probability of their joint occurrence, expressible through the terminal value and the running minimum. Reflection can help derive their joint distribution. The discussion outlines the approach but does not complete the integral or give a final price. Its setup is limited to the specified Brownian process and barrier arrangement; other dynamics or payoff details require a different calculation.
Key ideas
- The barrier condition and terminal payoff event are dependent under the stated process.
- The option value depends on the joint probability of finishing above the strike and crossing the lower barrier.
- Reflection principles can be used to study Brownian motion’s terminal value and running minimum jointly.
- Multiplying separate event probabilities would misprice the payoff unless independence were established.
Tags
Full text
# Fair value of a binary cash-or-nothing option with a barrier
# Fair value of a binary cash-or-nothing option with a barrier
I want to find the fair value of a European cash-or-nothing option that pays \$1 if $S_t>K$ and $S$ breached the level $M<0<K$, where $S$ is the risk-neutral process $dS_t=\sigma dW_t$.
My idea is to define a first passage time $\tau$ to level $M$ (since the other condition must be met anyway to get \$1 at time $T$) $(\tau=\min\{t;S_t=M\})$ and use the reflection principle of the Brownian motion to determine the probability density of $\tau$.
Integrating both sides of the SDE we find its solution $S_t=S_0+\sigma W_t$. Then, we applying the reflection principle and change of variable in integration $\nu=w/\sqrt{t} \Rightarrow d\nu=dw/ \sqrt{t}$:
\begin{align*} \mathbb{P}(\tau\leq t)&=\mathbb{P}(\tau\leq t,S_t\geq M)+\mathbb{P}(\tau\leq t,S_t\leq M) \\ & = 2\mathbb{P}(\tau\leq t,S_t\geq M) \\ &=2\mathbb{P}(S_t\geq M) \\ & = 2\int_{M}^{\infty}\frac{1}{\sqrt{2\pi t}}e^{-w^2/2t}dw \\ & = 2\int_{M/\sqrt{t}}^{\infty}\frac{1}{\sqrt{2\pi t}}e^{-\nu^2/2}d\nu \\ & = 2-2\Phi\left(\frac{M}{\sqrt{t}}\right) \end{align*}
The fair value of a standard cash-or-nothing option is $\mathbb{E}^\mathbb{Q}[\mathbb{I}_{\{S_t>K\}}]$. In this case, I think that we need to multiply that by $\mathbb{P}(\tau\leq t)$, i.e. the price of the cash-or-nothing option with barrier is:
$$\mathbb{E}^\mathbb{Q}[\mathbb{I}_{\{S_t>K\}}]\times\mathbb{P}(\tau\leq t)$$ Do you think this is correct?
## Answer by byouness (score 3, accepted)
https://quant.stackexchange.com/a/43390
As Daneel mentioned in his comment, you can't simply split your expectation of product into a product of two expecations as the two quantities are far from being independent...
Now, to answer your question w.r.t. how you could compute the expectation of the joint event of being in the money while having hit the barrier, you were right in using the reflexion principle. But I'd say you weren't ambitious enough :p
From the question, it appears your barrier is a down and in barrier.
Try using the reflexion principle to determine the joint law of $(S_t, \min_{s \leq t} S_s)$. Looking at the derivation you did in your question, this should be easy for you.
After you have done that, you can simply express the quantity that you want, which is: $$\mathbb{Q}\left( S_t>K, \min_{s \leq t} S_s < M \right)$$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.