【10】极大似然估计量的性质
【10】极大似然估计量的性质
(定理1) 设\(X_{(i)}(i=1,...,n)\sim N_p(\mu,\Sigma),(n>p)\),则(\(\mu,\Sigma\))'s MLE is:
(定理2) 若\(\overline{X},A\)分别为\(p\)元正态总体\(N_p(\mu,\Sigma)\)的样本均值向量,和样本离差阵,则:
- \(\overline{X}\sim N_p(\mu,\frac1n\Sigma)\);
- \(A=^d\sum_{t=1}^{n-1}Z_tZ_t'\),其中,\(Z_1,...,Z_{n-1}\)独立同\(N_p(0,\Sigma)\)分布;
- \(\overline{X},A\)相互独立;
- \(P\{A>0\}=1\Leftrightarrow n>p\).
讨论何时\(A\)为一个正定矩阵;
- 一元的一些结论:
if \(X_1,...X_n\) iid \(N(\mu,\sigma^2)\):
- \(\overline{X}\sim N(\mu,\frac{\sigma^2}{n})\);
- \(\frac{\sum_{i-1}^n(X_i-\overline{X})^2}{\sigma^2}\sim\chi^2(n-1)\);
- \(\overline{X}\)与\(s^2=\frac1{n-1}\sum_{i=1}^n(X-\overline{X})^2\)独立;
证明:
设 \(\Gamma\) 为 \(n\) 阶正交阵(不同行内积为 0 , 同行内积为 1),具有以下形式:
\[\Gamma= \left[ \begin{array}{ccc} \gamma_{11}&\dots&\gamma_{1n}\\ \vdots&&\vdots\\ \gamma_{(n-1),1}&\dots&\gamma_{(n-1),n}\\ \frac1{\sqrt{n}}&\dots&\frac1{\sqrt{n}} \end{array} \right]=(\gamma_{ij})_{n\times n} \]给 \(X\) 做线性变换 \(Z=\Gamma X\) 即:
\[Z=\left[ \begin{array}{c} Z_1'\\ \vdots\\ Z_n' \end{array} \right]=\Gamma \left[ \begin{array}{c} X_{(1)}'\\ \vdots\\ X_{(n)}' \end{array} \right]=\Gamma X \]则:
\[\begin{align} Z'=&X'\Gamma'\\\\ (Z_1,...,Z_n)=&(X_{(1)},...,X_{(n)})\Gamma'\\ Z_t=(X_{(1)},...,X_{(n)})&\left[ \begin{array}{c} \gamma_{t1}\\ \vdots\\ \gamma_{tn} \end{array} \right],(t=1,...,n) \end{align} \]\(Z_t\)为\(p\)维随机向量,而且是\(p\)维正态 r.v. \(X_{(1)}',...,X_{(n)}'\)的线性组合,故\(Z_t\)也是\(p\)维正态随机向量。且:
\[E(Z_t)=E(\sum_{i=1}^nr_{ti}X_{(i)})=\sum_{i=1}^nr_{ti}E(X_{(i)})=\mu\sum_{i=1}^nr_{ti}= \begin{cases} \mu\cdot(\frac1{\sqrt{n}}\sum_{i=1}^nr_{ti})\cdot\sqrt{n}&=0&当\,t\neq n\,时,\\ \mu\cdot n\frac1{\sqrt{n}}&=\sqrt{n}\mu&当\,t=n\,时\\ \end{cases} \]\[\begin{align} Cov(Z_{\alpha},Z_{\beta})=&E[(Z_{\alpha}-E(Z_\alpha))(Z_{\beta}-E(Z_{\beta}))']=\sum_{i=1}^n(r_{\alpha i}r_{\beta i})\Sigma= \begin{cases} O&\alpha\neq\beta,\\ \Sigma&\alpha=\beta \end{cases} \end{align} \]
- \(\overline{X}\sim N_p(\mu,\frac1n\Sigma)\);
\(Z_n=\frac1{\sqrt{n}}\sum_{\alpha=1}^nX_{(\alpha)}=\sqrt{n}\overline{X}\sim N_p(\mu\sqrt{n},\Sigma)\)
- \(A=^d\sum_{t=1}^{n-1}Z_tZ_t'\),其中,\(Z_1,...,Z_{n-1}\)独立同\(N_p(0,\Sigma)\)分布;
\(\sum_{\alpha=1}^nZ_\alpha Z_\alpha'=(Z_1,...,Z_n)\left(\begin{array}{c}Z_1'\\\vdots\\Z_{n}'\end{array}\right)=Z'Z=X'\Gamma'\cdot\Gamma X=\sum_{\alpha=1}^nX_\alpha X_\alpha'\)
于是:
\[\sum_{\alpha=1}^{n-1}Z_\alpha Z_\alpha'=\sum_{\alpha=1}^nX_\alpha X_\alpha'-Z_{n}Z_{n}'=\sum_{\alpha=1}^nX_\alpha X_\alpha'-n\overline{X}\overline{X}'=A \]
- \(\overline{X},A\)相互独立;
\[A=g(\sum_{t=1}^{n-1}Z_tZ_t')\\ \overline{X}=f(Z_n) \]而\(Z_i,Z_j\)相互独立(\(i\neq j\))时,则\(A、\overline{X}\)也相互独立。
- \(P\{A>0\}=1\Leftrightarrow n>p\).
\(B=(Z_1,...,Z_{n-1})\),记\(A=BB'\),因为\(A=BB'\),\(p\times(n-1)\)矩阵,显然\(rank(A)=rank(B)\),当\(A\)为正定矩阵时,\(rank(A)=p\),因此\(rank(B)=p\),故 \((n-1)\geq p\),即\(n>p\).
(无偏性)
故\(\hat{\Sigma}=\frac1nA\)不是\(\Sigma\)的无偏估计,应修正为:\(S=\frac1{n-1}A\).