Does the Markov property of a Random Variable allow us to say that the expected value of the future is independent of the past, given the present?

43 Views Asked by At

Given some probability space, say $(\Omega, \mathcal{F}, \mathbb{P})$, and the fact that $\mathcal{X}$ is $\mathbb{F}$-adapted, and $\mathcal{X}$ is Markov, then I know the following is true:

$\mathbb{P}(\mathcal{X_t} \in A | \mathcal{F}_s) = \mathbb{P}(X_t \in A | \mathcal{X}_s)$

...which follows directly from the fact that $\mathcal{X}_t - \mathcal{X}_s \mathrel{\unicode{x2AEB}} \mathcal{X}_s \, \forall s \le t$

But is it also true to say these two claims:

1) $\mathcal{X}_t \mathrel{\unicode{x2AEB}} \mathcal{F}_s | \mathcal{X}_s \, \forall s \le t$

2) $\mathbb{E}\mathcal{X}_t \mathrel{\unicode{x2AEB}} \mathcal{F}_s | \mathcal{X}_s \, \forall s \le t$

It seems to be true intuitively but I cannot derive mathematically.