PARTIALLY SUPERFLUOUS OBSERVATIONS
研究了一般线性回归模型中，给定第一个子样本时，第二个子样本对系数向量子集的最佳线性无偏估计是否部分冗余，给出了充要条件。
Necessary and sufficient conditions are established for the second subsample to be partially redundant given the first subsample for the best linear unbiased estimator of a subset of the coefficient vector of a general linear regression model.We thank the Co-editor for his very helpful comments on an earlier draft.