摘要
对增长曲线模型的回归系数提出了生成分估计,并证明了生成分估计优于最小二乘估计.进一步,对最小二乘估计的任一线性交换,给出其均方误差的一个无偏估计,并应用极小化均方误差的无偏估计的方法,给出了确定偏参数的公式.
This paper considers principal component estimate of regression coefficient in groWth curve model and proves that principal component estimate is superior to least squares estimate (LSE). Furthermore,the unbiased estimate of the mean square error(MSE) of any linear transformation for LSE is given. Using the method of minimixed unbiased estimate of the MSE of principal component estimate,the formala of determining biased parameter with principal component estimate is shown.
基金
武汉工业大学青年科学基金
关键词
增长曲线模型
回归系数
均方误差
主成分估计
growth curve model
regression coefficient
mean square error
principal component estimate