Let $\ X(t),t \ge 0$ be a Brownian motion process.

That is, $\ X(t)$ is a process with independent increments such that:

$$\ X(t) - X(s) \sim N(0,t-s), 0\le s \lt t $$
and $\ X(0)= 0$.

Derive the conditional distribution of $\ X(s), s \lt t $ conditional on $\ X(t) = B$ and state its mean and variance.

(I am pretty sure from looking online that the mean = $\ {Bs\over t} $ and variance = $\ {s(t-s) \over t} $ but I cannot derive the distribution to show this)


$X(t) \sim N(0,t)$ and $X(s) \sim N(0,s)$. $\operatorname{Var}(X(t) - X(s)) = \operatorname{Var}(X(t))+\operatorname{Var}(X(s))-2\operatorname{Cov}(X(t),X(s)) = t - s$ ==> $\operatorname{Cov}(X(t),X(s))=s$.

So $$\left( \begin{matrix} X(s)\\X(t) \end {matrix}\right) \sim N\left[\left( \begin{matrix} 0\\0 \end {matrix}\right),\left( \begin{matrix} s&s\\s&t \end {matrix}\right)\right]$$

Following $Y|X=x∼N\left(μ_Y+ρ\frac {σ_Y}{σ_X}(x−μ_X),σ_Y^2(1−ρ^2)\right)$, you get your results.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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