【FinE】Ito Integral习题(2)

习题

Q1：Generalized Geometric Brownian Motion

题面：
Let $S(t)$ be a positive stochastic process that satisfies
$$dS(t)=\alpha (t)S(t)dt+\sigma (t)S(t)dW(t)$$
where $\alpha (t)$ and $\sigma (t)$ are processes adapted to the Filtration $\mathcal{F}(t),t\ge 0$，compute $S(t)$ through computing $d\ln (S(t))$.
解析：
根据伊藤公式
$$\ln S_t=\ln S_0+{\int }_0^t{\sigma }_{s}d{w}_s+{\int }_0^t({\alpha }_s-\frac{1}{2}{\sigma }_s^2)ds$$
方程两边同时取$\exp$指数得到
$$S_t=S_0\exp ({\int }_0^t{\sigma }_{s}d{w}_s+{\int }_0^t({\alpha }_s-\frac{1}{2}{\sigma }_s^2)ds)$$

Q2：计算Brownian Motion高阶矩

题面：
Let $W(t)$ be a Brownian motion for $0\le t\le T$，Please find the expection $\mathbb{E}{W}^6(T)$.
解析：
伊藤积分
$$I(t)={\int }_0^t\Delta (s)dw(s)$$
期望
$$\mathbb{E}[I(t)]=0$$
方差
$$\mathbb{E}[I^2(t)]=\mathbb{E}[{\int }_0^t{\Delta }^2(u)du]={\int }_0^t\mathbb{E}[{\Delta }^2(u)]du$$
设$f(t,x)=x^6,f_t=0,f_x=6x^5,f_{xx}=30x^4$
由伊藤公式
$$df(t,W(t))=d{W}^6(t)=6W^5(t)dW(t)+\frac{1}{2}\ast 30W^4(t)dt=6W^5(t)dW(t)+15W^4(t)dt$$
在$[0,T]$上积分
$$W^6(T)-W^6(0)=\underset{\text{伊藤积分期望为0}}{\underbrace{{\int }_0^{T}6W^5(t)dW(t)}}+{\int }_0^{T}15W^4(t)dt$$
两边同时求数学期望
$$\mathbb{E}[W^6(T)]=15{\int }_0^T\mathbb{E}{W}^4(t)dt$$
根据性质，如果$X\sim \mathcal{N}(0,{\sigma }^2)，\mathbb{E}[X^4]=3{\sigma }^4$.
$\because W(t)\sim \mathcal{N}(0,t)\therefore \mathbb{E}[W^4(t)]=3t^2$
代入积分方程得到
$$\mathbb{E}[W^6(T)]=15T^3$$

Q3：Vasicek Model

题面：
The Vasicek interest rate stochastic differential equation is
$$dR(t)=(\alpha -\beta R(t))dt+\sigma dW(t)$$
where $\alpha ,\beta$ and $\sigma$ are positive constants. Please compute $R(t)$ through computing $d(e^{\beta t}R(t))$.
解析：
令$f(t,x)=e^{\beta t}x,f_t=x\beta e^{\beta t},f_x=e^{\beta t},f_{xx}=0$
计算微分并代入$R(t)$表达式得
$$\begin{array}{rl}d(e^{\beta t}R(t))& =d(f(t,R(t)))=R(t)\beta e^{\beta t}dt+e^{\beta t}dR(t)\\ & =\alpha e^{\beta t}dt+\sigma e^{\beta t}dW(t)\end{array}$$
在$[0,t]$上对方程两边进行积分
$$e^{\beta t}(R(t))-R(0)={\int }_0^t\alpha e^{\beta s}ds+{\int }_0^t\sigma e^{\beta s}dW(s)$$
移项积分求解得到
$$R(t)=e^{-\beta t}R(0)+\frac{\alpha }{\beta }(1-e^{-\beta t})+{\int }_0^t\sigma e^{-\beta (t-s)}dW(s)$$
可以计算一些性质，利用伊藤积分期望为0
$$\mathbb{E}[R(t)]=e^{-\beta t}R(0)+\frac{\alpha }{\beta }(1-e^{-\beta t})$$
计算极限
$$\underset{t\to \infty }{\lim }\mathbb{E}[R(t)]=\frac{\alpha }{\beta }$$
可以看出$R(t)$具有均值回归的特性.

Q4：Stratonovich Integral

题面：
Let $W(t),0\le t\le T$ be a Brownian motion. Let $T$ be a fixed positive number and let Π = { t 0 , t 1 , … , t n } \Pi=\{t_0, t_1, \dots, t_n\} Π={t0​,t1​,…,tn​} be a partition of $[0,T]$. For each $j$, define $t_j^{\ast }=\frac{t_j+t_{j+1}}{2}$ to be the midpoint of the internal $[t_j,t_{j+1}]$.

(1) Define the half-sample quadratic variation corresponding to $\Pi$ to be
$$Q_{\Pi /2}=\sum _{j=0}^{n-1}[W(t_j^{\ast })-W(t_j){]}^2$$
Show that $Q_{\Pi /2}$ has limit $\frac{1}{2}T$ as $\Vert \Pi \Vert \to 0$
解析：
计算$\mathbb{E}[Q_{\Pi /2}]$
$$\mathbb{E}[Q_{\Pi /2}]=\sum _{j=0}^{n-1}\mathbb{E}[W(t_j^{\ast })-W(t_j){]}^2$$
令$D_j=W(t_j^{\ast })-W(t_j)$

$$D_j=W(t_j^{\ast })-W(t_j)\sim \mathcal{N}(0,t_j^{\ast }-t_j)=\mathcal{N}(0,\frac{t_{j+1}-t_j}{2})$$
$\therefore$
$$\mathbb{E}[Q_{\Pi /2}]=\sum _{j=0}^{n-1}\mathbb{E}[D_j^2]$$
$\therefore$
$$\mathbb{V}[D_j]=\mathbb{E}[D_j^2]-\mathbb{E}[D_j{]}^2=\frac{t_{j+1}-t_j}{2}$$
$\therefore$
$$\mathbb{E}[Q_{\Pi /2}]=\sum _{j=0}^{n-1}\frac{t_{j+1}-t_j}{2}=\frac{t_n-t_0}{2}=\frac{T}{2}$$
计算方差，根据布朗运动的增量独立性
$$\mathbb{V}[Q_{\Pi /2}]=\mathbb{V}[\sum _{j=0}^{n-1}{D}_j^2]=\sum _{j=0}^{n-1}\mathbb{V}[D_j^2]$$
$\because$
$$\mathbb{V}[D_j^2]=\mathbb{E}[D_j^4]-{\mathbb{E}}^2[D_j^2]=\frac{3}{4}(t_{j+1}-t_j{)}^2-\frac{1}{4}(t_{j+1}-t_j{)}^2=\frac{1}{2}(t_{j+1}-t_j{)}^2$$
$\therefore$
$$\mathbb{V}[Q_{\Pi /2}]=\frac{1}{2}\sum _{j=0}^{n-1}(t_{j+1}-t_j{)}^2$$
当$\max (t_{j+1}-t_j)=\Vert \Pi \Vert \to 0$时
$$\mathbb{V}[Q_{\Pi /2}]=\frac{1}{2}\sum _{j=0}^{n-1}(t_{j+1}-t_j)(t_{j+1}-t_j)\le \frac{1}{2}\Vert \Pi \Vert \sum _{j=0}^{n-1}(t_{j+1}-t_j)=\frac{1}{2}\Vert \Pi \Vert T\to 0$$
(2) Define the Stratonovich Integral of $W(t)$ with respect to $W(t)$ to be
$${\int }_0^{T}W(t)\circ dW(t)=\underset{\Vert \Pi \Vert \to 0}{\lim }\sum _{j=0}^{n-1}W(t_j^{\ast })[W(t_{j+1})-W(t_j)]$$
证明
$${\int }_0^{T}W(t)\circ dW(t)=\frac{1}{2}{W}^2(T)$$
解析：
考虑从伊藤积分(区间补全)凑出Stratonovich Integral
Ito Intrgral
$$I={\int }_0^{T}W(t)dW(t)=\sum _{j=0}^{n-1}W(t_j)[W(t_j^{\ast })-W(t_j)]+\sum _{j=0}^{n-1}W(t_j^{\ast })[W(t_{j+1})-W(t_j^{\ast })]$$
Stratonovich Integral
$$\begin{array}{rl}S& =I-2\sum _{j=0}^{n-1}W(t_j)W(t_j^{\ast })+\sum _{j=0}^{n-1}{W}^2(t_j)+\sum _{j=0}^{n-1}{W}^2(t_j^{\ast })\\ & =I+\sum _{j=0}^{n-1}[W(t_j)-W(t_j^{\ast }){]}^2\\ & \stackrel{由(1)结果}{=}I+\frac{T}{2}\end{array}$$
$\therefore$
$${\int }_0^{T}W(t)\circ dW(t)={\int }_0^{T}W(t)dW(t)+\frac{T}{2}$$
求伊藤积分
令$f(t,x)=x^2,f_t=0,f_x=2x,f_{xx}=2$
$$df(t,W(t))=d[W^2(t)]=2W(t)dW(t)+\frac{1}{2}\ast 2\ast dt=2W(t)dW(t)+dt$$
$\therefore$
$$\begin{array}{rl}& {\int }_0^T{W}^2(t)dW(t)={\int }_0^{T}2W(t)dW(t)+{\int }_0^{T}dt\\ & W^2(T)-W^2(0)=2{\int }_0^{T}W(t)dW(t)+{\int }_0^{T}dt\\ & {\int }_0^{T}W(t)dW(t)=\frac{W^2(T)-T}{2}\end{array}$$
代入计算可以得到
$${\int }_0^{T}W(t)\circ dW(t)={\int }_0^{T}W(t)dW(t)+\frac{T}{2}=\frac{W^2(T)}{2}$$

参考资料

Ito Integral习题 part 2 Jerry Xu