Applying Ito's lemma on $E[U \exp(i\sum_j \int_0^\infty \alpha^j dB^j)|\mathcal{F}_t]$ on the natural filtration of Brownian motion

21 Views Asked by At

This is from the following post : https://almostsuremath.com/2010/05/25/the-martingale-representation-theorem/#comment-12778

Let $B=(B^1, \dots, B^d)$ be a $d$-dimensional Brownian motion defined on the natural filtration $(\mathcal{F}_t)_{t\ge 0})$. Let $\alpha^j$ be deterministic processes satisfying $\int_0^\infty (\alpha_t^j)^2dt<\infty$. $U$ a $\mathcal{F}_0$ measurable random variable. And $i$ is the imaginary number.

Let $$Z = U \exp(i\sum_j \int_0^\infty \alpha^j dB^j)$$ and then we can write $$E[Z|\mathcal{F}_t]=U\exp(i \sum_j \int_0^t \alpha^j dB^j - \frac{1}{2}\sum_j \int_t^\infty (\alpha_s^j)^2 ds).$$

Now if we set $M_t = E[Z|\mathcal{F}_t]$,

how do we use the Ito's lemma here to get $$M_t = M_0 + i\sum_j \int_0^t M\alpha^j dB^j?$$

Moreover, if we set $\xi^j = iM\alpha^j$, then how do we ensure that $\int_0^t (\xi_s^i)^2 ds<\infty$ almost surely for each $t\ge 0$?

I would greatly appreciate if anyone could explain these two questions to me.

enter image description here