约束线性回归模型回归系数的广义条件岭估计

Generalized Conditional Ridge-type Estimation of Regression Coefficient in Restricted Linear Regression Model

对齐次等式约束线性回归模型回归系数的岭估计进行推广,得出广义条件岭估计.证明了在一定的条件下,在均方误差矩阵、均方误差及广义均方误差意义下广义条件岭估计都优于约束最小二乘估计,并且证明了广义条件岭估计是回归系数的可容许估计.

We propose the generalized conditional ridge-type estimation of regression coefficient in restricted linear regression model. We show that is restricted admissible and superior to the restricted least squares estimation in terms of mean squares error and mean squares error matrix.