Advances and Applications in Statistics
Volume 3, Issue 2, Pages 135 - 143
(August 2003)
|
|
MIXED-EFFECTS LINEAR MODELS BASED ON ELLIPTICALLY CONTOURED DISTRIBUTION ASSUMPTIONS
Karan P. Singh (U. S. A.), Alfred A. Bartolucci (U. S. A.) and Shimin Zheng (U. S. A.)
|
Abstract: In this paper
we extend the random coefficient models based on
the normal assumptions to the random coefficient
models based on the ECD assumptions. We
derived the estimators of the fixed parameters
in the random coefficient models under the ECD
assumptions. We also derived the formulae for
computing the best linear unbiased predictors of
the random intercepts and random slopes in the
random regression model for the longitudinal
data and the variances of these estimates. |
Keywords and phrases: random regression, dispersion matrix, random parameters, best linear unbiased estimates, best linear unbiased predictor. |
|
Number of Downloads: 368 | Number of Views: 1215 |
|