夜空中最亮的星
1- Dirichlet 积分
设\(I(a)=\frac1\pi\int_0^{+\infty}\frac{\sin{at}}{t}dt\),则有:
\[
I(a)=
\begin{cases}
\frac12&\text{a>0}\\
0&a=0\\
-\frac12&a<0
\end{cases}
\]
为了证明 \(Dirichlet\ 积分\),我们先证明\(\int_0^{+\infty}\frac{\sin{x}}{x}dx=\frac\pi2\)
\[
\begin{align}
设\ \frac1x=&\int_0^{+\infty}e^{-xs}ds\\
\int_0^T\frac{\sin{x}}{x}dx=&\int_0^{T}(\sin{x}{\int_0^{+\infty}e^{-xs}ds)}dx\\
=&\int_0^{+\infty}({\int_0^{T}\sin{x}\ e^{-xs}dx)}ds\\
=&\int_0^{+\infty}[\frac{1}{1+s^2}-\frac{s\cdot\sin T+T\cdot\cos{T}}{s^2+T^2}e^{-s}]ds\\
=&\frac\pi2-\int_0^{+\infty}\frac{s\cdot\sin T+T\cdot\cos{T}}{s^2+T^2}e^{-s}ds\\
(\because&\lim_{s\to\infty}\frac{s\cdot\sin T+T\cdot\cos{T}}{s^2+T^2}=0)\\
=&\frac\pi2\\
则I(a)=&\frac1\pi\int_0^{+\infty}\frac{\sin{at}}{t}dt\\
=&a\cdot\frac1a\cdot\frac1\pi\int_0^{+\infty}\frac{\sin{at}}{at}d(at)\\
=&\frac1\pi\cdot\frac\pi2\\
=&\frac12
\end{align}
\]
此积分的重要意义是可以将符号转化为积分表达式!
3-特征函数
设\(X\)是随机变量,\(f(x)\)为概率密度函数,则特征函数为:
\[
\varphi_X(t)=\int_{-\infty}^\infty e^{itx}f(x)dx
\]
由:\(e^{itx}=\sum_{n=0}^\infty\frac{(itx)^n}{n!}\),则有:
\[
\begin{align}
\varphi_X(t)=&\int_{-\infty}^\infty e^{itx}f(x)dx\\
=&\int_{-\infty}^\infty (\sum_{n=0}^\infty\frac{(itx)^n}{n!})f(x)dx\\
=&\sum_{n=0}^\infty\int_{-\infty}^\infty(\frac{(itx)^n}{n!})f(x)dx\\
=&\sum_{n=0}^\infty\frac{(it)^n}{n!}\int_{-\infty}^\infty x^nf(x)dx\\
=&\sum_{n=0}^\infty\frac{(it)^n}{n!}E(x^n)\\
\varphi_X(t)=&e^{it}\cdot E(x^n)
\end{align}
\]
特征函数的反演公式:
\[
f(x)=\frac1{2\pi}\int_{-\infty}^{\infty}e^{-itx}\varphi(t)dt
\]
4-特征函数性质
设\(\{f(x)=P(x\leq X)\ |\ E(X)=\mu,Var(X)=\sigma^2\}\),其特征函数的本质为概率密度函数的傅里叶变换:
\[
\varphi_X(t)=\int_{-\infty}^\infty e^{itx}f(x)dx
\]
而特征函数的 k 阶导数与 k 阶矩之间有密切联系:
\[
\varphi^{(k)}_X(t)=[\int_{-\infty}^\infty e^{itx}f(x)dx]^{(k)}=\int_{-\infty}^\infty (e^{itx})^{(k)}f(x)dx
=\int_{-\infty}^\infty (ix)^{k}e^{itx}f(x)dx
\]
于是,令\(t=0\):
\[
\begin{align}
\varphi_X(0)&=\int_{-\infty}^\infty f(x)dx=1\\
\varphi^{'}_X(0)&=\int_{-\infty}^\infty (ix)e^0f(x)\ dx\\
&=i\int_{-\infty}^\infty xf(x)\ dx\\
&=i\cdot E(x)\\
\varphi^{''}_X(0)&=\int_{-\infty}^\infty (ix)^2e^0f(x)\ dx\\
&=-\int_{-\infty}^\infty x^2f(x)\ dx\\
&=-(\int_{-\infty}^\infty (x-\mu)^2f(x)\ dx+\int_{-\infty}^\infty(2\mu)xf(x)\ dx-\mu^2\int_{-\infty}^\infty f(x)\ dx)\\
&=-\sigma^2-\mu^2
\end{align}
\]
特征函数的乘法性质:
\[
\varphi_{X+Y}(t)=\varphi_{X}(t)\varphi_{Y}(t)
\]
若\(X\thicksim N(0,1)\),则其概率密度函数为:
\[
f(x)=\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}
\]
其对应的特征函数为:
\[ \begin{align} \varphi_X(t)=&\int_{-\infty}^\infty e^{itx}(\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}})dx=e^{-\frac{t^2}{2}} \end{align} \]
\[ \begin{align} \varphi_X(t)=&\int_{-\infty}^\infty e^{itx}(\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}})dx\\ \varphi'_X(t)=&\frac{1}{\sqrt{2\pi}}\int_{-\infty}^\infty(-xe^{-\frac{x^2}{2}}+ixe^{-\frac{x^2}{2}})e^{itx}dx\\ =&\frac{1}{\sqrt{2\pi}}\int_{-\infty}^\infty ixe^{-\frac{x^2}{2}}e^{itx}dx\\ =&\frac{i}{\sqrt{2\pi}}\int_{-\infty}^\infty e^{itx}d(-e^{-\frac{x^2}{2}})\\ =&\frac{i}{\sqrt{2\pi}}e^{(itx-\frac{x^2}{2})}|_{-\infty}^{+\infty} -\frac{i}{\sqrt{2\pi}}\int_{-\infty}^\infty -e^{-\frac{x^2}{2}}d(e^{itx})\\ =&\frac{i}{\sqrt{2\pi}}e^{(itx-\frac{x^2}{2})}|_{-\infty}^{+\infty} -\frac{i}{\sqrt{2\pi}}\int_{-\infty}^\infty ite^{itx}(-e^{-\frac{x^2}{2}})dx\\ =&-\frac{t}{\sqrt{2\pi}}\int_{-\infty}^\infty e^{itx}(e^{-\frac{x^2}{2}})dx\\ =&-t\varphi(t) \end{align} \]
于是我们又获得一个重要的微分方程:
\[
\frac{d\varphi(t)}{dt}=-t\cdot\varphi(t)
\]
变形得到:
\[
\begin{align}
\frac{d\varphi(t)}{\varphi(t)}&=-tdt\\
\int\frac{d\varphi(t)}{\varphi(t)}&=-\int tdt\\
\ln{\varphi(t)}&=-\frac{t^2}{2}+C\\
\varphi(t)&=C_0\cdot e^{-\frac{t^2}{2}}\\
(\because\varphi(0)=1,带入公式&得到C_0=e^0=1,于是有)\\
\varphi(t)&=e^{-\frac{t^2}{2}}\\
\end{align}
\]
也就是说当概率密度函数为\(f(x)=\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}\)时,特征函数为:\(\varphi(t)=e^{-\frac{t^2}{2}}\)。
5-中心极限定理
设\(X_1,X_2,\dots,X_n\)为n个独立同分布随机变量,\(X_i\thicksim N(\mu,\sigma^2)\),不妨设\(Y_i=X_i-\mu,\)那么\(Y_i\thicksim N(0,\sigma^2)\),设\(Y_i\)的特征函数为\(\varphi(t)\),
设随机变量\(\eta=\frac{Y_1+Y_2+\dots+Y_n}{\sigma\sqrt{n}}\),由\(\varphi_{X+Y}(t)=\varphi_{X}(t)\varphi_{Y}(t)\),则\(\eta\)的特征函数为:
\[
[\varphi(\frac{t}{\sqrt{n}\sigma})][\varphi(\frac{t}{\sqrt{n}\sigma})]\dots[\varphi(\frac{t}{\sqrt{n}\sigma})]=[\varphi(\frac{t}{\sqrt{n}\sigma})]^n
\]
将\(\varphi(\frac{t}{\sqrt{n}\sigma})\)在0点\(Taylor\)展开,
\[
\begin{align}
\varphi(\frac{t}{\sqrt{n}\sigma})
&=\varphi(0)+\varphi'(0)(\frac{t}{\sqrt{n}\sigma})+\frac{\varphi''(0)}{2!}(\frac{t}{\sqrt{n}\sigma})^2+o((\frac{t}{\sqrt{n}\sigma})^2)\\
&=1+\frac{i\mu t}{\sqrt{n}\sigma}-\frac{(\sigma^2+\mu)t^2}{2n\sigma^2}+o((\frac{t}{\sqrt{n}\sigma})^2)\\
&=1-\frac{t^2}{2n}+o((\frac{t}{\sqrt{n}\sigma})^2)\\
\therefore[\varphi(\frac{t}{\sqrt{n}\sigma})]^n&=[1-\frac{t^2}{2n}+o((\frac{t}{\sqrt{n}\sigma})^2)]^n\\
&=[1-\frac{t^2}{2n}+o((\frac{t}{\sqrt{n}\sigma})^2)]^{(-\frac{2n}{t^2})(-\frac{t^2}{2})}\\
\lim_{n\to\infty}[\varphi(\frac{t}{\sqrt{n}\sigma})]^n&=e^{-\frac{t^2}{2}}
\end{align}
\]
因此\(\eta\thicksim N(0,1)\)服从标准正态分布。
来源:https://www.cnblogs.com/rrrrraulista/p/12310512.html