摘要
本文主要从判决理论的观点出发，给出使用无偏估计的条件，求参数的无偏估计，在损失为严凸的情况下，先找一个完全统计量，再找任一无偏估计＾ｇ（ Ｘ） ，计算条件期望 Ｅ（＾ｇ｜ Ｔ） 即可。
By using the theory of judgement as a starting point the condition of unbiased estimation was offered so as to obtain the unbiased estimation of the parameter when loss was strictly convex . First of all, perfect statistics (X)was found, and then the conditional expectation E(｜T) was caculated.
关键词
无偏估计
统计量
样本
unbiased estimation, statistics, sample.
