返回题库

几何布朗运动:求 dX(t)

Geometric Brownian Motion

专题
Finance / 金融
难度
L6

题目详情

W(t)W(t) 为标准布朗运动,

X(t)=eσW(t)+(μσ22)t,X(t)=e^{\sigma W(t)+\left(\mu-\frac{\sigma^2}{2}\right)t},

其中 μ,σ\mu,\sigma 为常数。求 dX(t)dX(t)

Let W(t)W(t) be a standard Brownian motion. Consider the stochastic process defined by:

X(t)=eσW(t)+(μσ22)tX(t) = e^{\sigma W(t) + \left(\mu - \frac{\sigma^2}{2}\right)t}

where μ\mu and σ\sigma are constants. Find dX(t)dX(t), the differential of the process X(t)X(t).

解析

X(t)=f(t,W(t))X(t)=f(t,W(t)) 用伊藤公式。

可得

ft=(μσ22)X(t),fW=σX(t),2fW2=σ2X(t).\frac{\partial f}{\partial t}=\left(\mu-\frac{\sigma^2}{2}\right)X(t),\quad \frac{\partial f}{\partial W}=\sigma X(t),\quad \frac{\partial^2 f}{\partial W^2}=\sigma^2 X(t).

代入伊藤公式:

dX(t)=(ft+122fW2)dt+fWdW(t)=μX(t)dt+σX(t)dW(t).dX(t)=\left(\frac{\partial f}{\partial t}+\frac12\frac{\partial^2 f}{\partial W^2}\right)dt+\frac{\partial f}{\partial W}dW(t) =\mu X(t)dt+\sigma X(t)dW(t).

Original Explanation

To apply Itô’s Lemma to X(t)X(t), we treat XX as a function of tt and W(t)W(t), i.e., X(t)=f(t,W(t))X(t) = f(t, W(t)).

Let:
f(t,W(t))=eσW(t)+(μσ22)tf(t, W(t)) = e^{\sigma W(t) + \left(\mu - \frac{\sigma^2}{2}\right)t}

First, compute the partial derivatives:

  • Time derivative: ft=(μσ22)eσW(t)+(μσ22)t\frac{\partial f}{\partial t} = \left( \mu - \frac{\sigma^2}{2} \right) e^{\sigma W(t) + \left(\mu - \frac{\sigma^2}{2}\right)t}

  • First derivative with respect to WW: fW=σeσW(t)+(μσ22)t\frac{\partial f}{\partial W} = \sigma e^{\sigma W(t) + \left(\mu - \frac{\sigma^2}{2}\right)t}

  • Second derivative with respect to WW: 2fW2=σ2eσW(t)+(μσ22)t\frac{\partial^2 f}{\partial W^2} = \sigma^2 e^{\sigma W(t) + \left(\mu - \frac{\sigma^2}{2}\right)t}

Now, plug into Itô’s Lemma:

dX(t)=(ft+122fW2)dt+fWdW=((μσ22)+12σ2)X(t)dt+σX(t)dW(t)=μX(t)dt+σX(t)dW(t)\begin{aligned} dX(t) &= \left( \frac{\partial f}{\partial t} + \frac{1}{2} \frac{\partial^2 f}{\partial W^2} \right) dt + \frac{\partial f}{\partial W} dW \\ &= \left( \left(\mu - \frac{\sigma^2}{2}\right) + \frac{1}{2} \sigma^2 \right) X(t) dt + \sigma X(t) dW(t) \\ &= \mu X(t) dt + \sigma X(t) dW(t) \end{aligned}

Final Answer:
dX(t)=μX(t)dt+σX(t)dW(t)dX(t) = \mu X(t) \, dt + \sigma X(t) \, dW(t)