-1
$\begingroup$

I was working on numerical solution of this equation (by block pulse). $$x(t)=1+\int_{0}^{t}s^2x(s)ds+\int_{0}^{t}sx(s)dB(s)\\t \in[0,\frac{1}{2}]$$B(t) is standard brownian motion. Author of the paper said :the exact solution of this equation is $$ x(t)=e^{\frac{t^3}{6}+\int_{0}^{t}sdB(s)}$$ I tried to check the exact solution into it ,.I tried to find the exact solution ,but I can't.please help me to understand how obtain this exact solution (or check into equation).

$\endgroup$

1 Answer 1

1
$\begingroup$

Let $Y_t = \int_0^ts\,dB_s$. Note that $Y$ is a martingale and hence a semimartingale. Then consider the process $X_t = f(t,Y_t)$ with $f(t,x) = e^{\frac{t^3}{6}+x}$. Since the partial derivatives $f_t,f_x,f_{xx}$ all exist and are continuous, Ito's lemma applies in this situation. $$f(t,Y_t) = f(0,Y_0) + \int_0^tf_t(s,Y_s)\,ds + \int_0^tf_x(s,Y_s)\,dY_s + \frac{1}{2}\int_0^tf_{xx}(s,Y_s)\,d[Y,Y]_s$$ It is easy to compute that $[Y,Y]_t = \int_0^ts^2\,ds = \frac{t^3}{3}$. Furthermore, $dY_s = s\,dB_s$ and $X_0 = f(0,Y_0) = 1$. Putting everything together $$X_t = X_0 + \frac{1}{2}\int_0^ts^2X_s\,ds + \int_0^tsX_s\,dB_s + \frac{1}{2}\int_0^ts^2X_s\,ds = 1 + \int_0^tsX_s\,dB_s + \int_0^ts^2X_s\,ds$$

To show $[Y,Y]_t = \int_0^ts^2\,ds$, I used the following result.

For semimartingales $M,N$ $$[\int G_sdM_s,\int H_sdN_s]_t = \int_{(0,t]}G_sH_s\,d[M,N]_s$$

$\endgroup$
4
  • $\begingroup$ I had a type,I fix it ,$\frac{t^3}{3} \to \frac{t^3}{6}$ but do not make a big difference .Thank You .Can you explain ( abit more)how to compute $[y,y]t$ ?--do you use Ito Isometery formula ? $\endgroup$
    – Khosrotash
    Aug 18, 2015 at 8:56
  • $\begingroup$ @Khosrotash You are welcome. I edited my answer. Ito isometry is not relevant here. $\endgroup$
    – Calculon
    Aug 18, 2015 at 9:04
  • $\begingroup$ Ok.I got it . thank you very much $\endgroup$
    – Khosrotash
    Aug 18, 2015 at 9:37
  • $\begingroup$ @Khosrotash No problem. If my post answers your question, I would appreciate it if you could accept it. $\endgroup$
    – Calculon
    Aug 18, 2015 at 17:49

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .