【学习笔记】数理统计习题六

Q1: Let $X_1,\dots ,X_n$ be an iid sample of Poisson distribution with parameter $\lambda$. Which of the following are unbiased estimates of $\lambda$? ( $ABC$ ) 多选

A. $\bar{X}$

B. $S_n^{\ast 2}$

C. $(\bar{X}+S_n^{\ast 2})/2$

D. $S_n^2$

Q2: Let $X_1,\dots ,X_n$ be an iid sample of $N(\mu ,{\sigma }^2)$, where $\mu ,\sigma$ are unknown parameters. Let $T_k=k{\sum }_{i=1}^n(X_i-\bar{X}{)}^2$ be an estimator of ${\sigma }^2$. Particularly, when $k=1/n$, $T_k=S_n^2$, and when $k=1/(n-1)$, $T_k=S_n^{\ast 2}$. Find a value of $k$ such that $T_k$ is the most efficient one by taking account of MSE.

解： 因为$T_k/(k{\sigma }^2)\sim {\chi }^2(n-1)$，我们有$$E[T_k]=k(n-1){\sigma }^2,Var[T_k]=2k^2(n-1){\sigma }^4$$于是$$MSE(T_k)=(E[T_k]-{\sigma }^2{)}^2+Var[T_k]=[(n^2-1)k^2-2(n-1)k+1]{\sigma }^4$$容易得到当$k=1/(n+1)$时$MSE(T_k)$最小，估计量最有效

Q3: Let $X_1,\dots ,X_n$ be a simple random sample taken from the density

$$f(x;\theta )=\frac{2x}{{\theta }^2},0\le x\le \theta .$$

1. Find an expression for ${\hat{\theta }}_L$, the maximum likelihood estimator (MLE) for $\theta$.

   解： 似然函数为 L ( θ ) = ∏ i = 1 n 2 x i θ 2 = 2 n θ 2 n ( ∏ i = 1 n x i ) 1 { x ( n ) ≤ θ } L(\theta)=\displaystyle\prod_{i=1}^n\frac{2x_i}{\theta^2}=\frac{2^n}{\theta^{2n}}(\displaystyle\prod_{i=1}^nx_i)1\{x_{(n)}\le\theta\} L(θ)=i=1∏n​θ22xi​​=θ2n2n​(i=1∏n​xi​)1{ x(n)​≤θ}为了使似然函数取得最大，我们需要选择$\theta \ge x_{(n)}$，使得$L(\theta )=A{\theta }^{-2n}$，其中$A=2^n{\prod }_{i=1}^n{x}_i$不依赖于$\theta$，因此最大似然估计值为$\hat{\theta }=x_{(n)}$

2. Find an expression for ${\hat{\theta }}_M$, the method of moments estimator for $\theta$.

   解： $${\mu }_1=E(X)={\int }_0^{\theta }\frac{2x^2}{{\theta }^2}dx=\frac{2}{3}\theta$$ 以$A_1$代替${\mu }_1$，
   A 1 = 1 n ∑ i = 1 n X i = X ˉ = 2 3 θ A_1=\frac{1}{n}\displaystyle\sum_{i=1}^nX_i=\bar{X}=\frac{2}{3}\theta A1​=n1​