Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $W$ be a Wiener process, $r,\sigma \in \mathbb{R}_+$ and $S(T) = S(t)e^{(r-\frac12 \sigma^2)(T-t) + \sigma(W(T)-W(t))}$.

I want to evaluate $$A:=E[e^{- \frac12 \sigma^2 (T-t) - \sigma(W(T)-W(t)) } * \mathbf{1}_{\{S(T)>1.5\}}|\mathscr{F}_t].$$

You will see that the left-hand factor of the expectation is an exponential martingale of the form $\mathcal{E}(-\sigma W)_t$.

Question: What can I do to progress further? I feel like I can't go:

$$A = E[e^{- \frac12 \sigma^2 (T-t) - \sigma(W(T)-W(t)) }|\mathscr{F}_t] * E[\mathbf{1}_{\{S(T)>1.5\}}|\mathscr{F}_t]$$

because $W(T)$ appears both inside the $e^{...}$ and inside the $\mathbf{1}_{\{...\}}$, making them dependent.

share|cite|improve this question
Have tried Girsanov Theorem ? – TheBridge Nov 24 '12 at 8:08
@TheBridge I am not allowed to change probability measure in this question (they specifically told us so). – Jase Nov 24 '12 at 8:31
But $[S(T)\gt0]=[S(t)\gt0]$, no? Which would imply that the event belongs to $\mathcal F_t$, and make the answer trivial. – Did Nov 24 '12 at 11:58
@did I apologize for the typo. – Jase Nov 24 '12 at 12:16
Jase: Just out of curiosity, who is asking you to solve this? – Did Nov 24 '12 at 19:44
up vote 2 down vote accepted

Since $W(T)-W(t)$ is independent of $\mathcal F_t$, $A=\mathbb E(X\cdot\mathbf 1_{S(t)Y\gt x}\mid\mathcal F_t)$ for $x=1.5$ and for some random variables $X$ and $Y$ which are independent of $\mathcal F_t$. Hence $A=\mathbb E(X)\cdot G(S(t))$ where $G(z)=\mathbb P(zY\gt x)$.

To go further, one can note that $X=\exp(- \frac12 \sigma^2 (T-t) - \sigma(W(T)-W(t)))$ hence $\mathbb E(X)=1$, and that $Y=\exp((r- \frac12\sigma^2) (T-t)-\sigma\sqrt{T-t}Z)$, where $Z$ is standard normal, hence $$ G(z)=\Phi\left(\frac{\log z-\log x+(\frac12\sigma^2-r) (T-t)}{\sigma\sqrt{T-t}}\right), $$ or some similarly ugly and mostly useless formula.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.