已知

• $X\sim X(\theta_{1},\dots,\theta_{n})$

则

1. 用参数表示 $1$ ~ $n$ 阶总体原点矩 $$\left{\begin{align}\mu_{1} & =\mu_{1}(\theta_{1},\dots,\theta_{n}) \ \mu_{2} & =\mu_{2}(\theta_{1},\dots,\theta_{n}) \ & \vdots \ \mu_{n} & =\mu_{n}(\theta_{1},\dots,\theta_{n}) \end{align}\right.$$
2. 解得参数 $$\left{\begin{align}\theta_{1} &=\theta_{1}(\mu_{1},\dots,\mu_{n})\ \theta_{2} &=\theta_{2}(\mu_{1},\dots,\mu_{n}) \ & \vdots \ \theta_{n} &=\theta_{n}(\mu_{1},\dots,\mu_{n}) \end{align}\right.$$
3. 用样本原点矩 $A_{k}$ 代替总体原点矩 $\mu_{k}$ 得到估计量 $$\hat{\theta}{i}=\theta{i}(A_{1},\dots,A_{n})$$