Linear regression with randomly dispersed parameters
暂无分享,去创建一个
SUMMARY Consider a collection of individual linear regression models, in which each individual parameter vector is independently drawn from a common multivariate normal distribution and is fixed over successive observations on that individual. Maximum likelihood estimators of the mean and dispersion of the parameters and of the disturbance variance are derived. These estimators yield empirical Bayes estimators of the individual parameter vectors. The properties of the estimators are exhibited in the case where the parameter dispersion is known.
[1] A. C. Aitken. IV.—On Least Squares and Linear Combination of Observations , 1936 .
[2] P. A. V. B. Swamy,et al. Statistical Inference in Random Coefficient Regression Models , 1971 .
[3] P. R. Fisk,et al. Models of the Second Kind in Regression Analysis , 1967 .
[4] C. Hildreth,et al. Some Estimators for a Linear Model With Random Coefficients , 1968 .
[5] Calyampudi Radhakrishna Rao,et al. Linear Statistical Inference and its Applications , 1967 .