摘要
对增长曲线模型的回归系数提出了分组压缩主成份估计，并证明了分组压缩主成份估计优于最小二乘估计。进一步讨论了它的优良性。
This paper considers the classification compression principal component estimate of regression coefficient in growth curve model and proves that it is superior to least squares estimate.Furthermore,its optimal properties are shown and the method of aetermining biased parameter is also given.
关键词
增长曲线模型
回归系数
主成份估计
均方误差
growth curve model
regression coefficient
classification compression principal component estimate
mean square error.
