回归系数一类线性估计的小样本性质

Small sample properties of a kind of linear estimation in linear regression model

提出了线性模型中回归系数的一类线性估计.在均方误差矩阵(MSEM)准则和PitmanCloseness(PC)准则下,研究了这类线性估计相对于最小二乘(LS)估计的优良性.最后,讨论了当设计阵为非列满秩时,回归系数的可估函数的一类线性估计的优良性.

A kind of linear estimator is derived in linear regression model. The superiorities of the linear estimator over least square （LS） estimator are studied in terms of the mean square error matrix （MSEM） criterion and Pitman closeness （PC） criterion. Finally, the superiorities of the linear estimator of estimable function for regression coefficients are discussed in non-full rank case.