Likelihood ratio tests for linear hypotheses of parameters based on the maximum likelihood method in a hierarchical growth curve model
by
Takashi Kanda
Hiroshima Institute of Technology, JAPAN

In this paper we consider an extended growth curve model with hierarchical type of within-individuals design matrices. For an observation matrix Y, the model is defined by using known between-individuals design and within-individuals design matrices. It is assumed that the rows of Y are independently distributed as a p-variate normal distribution having unknown positive definite covariance matrix. We derive LR tests for some linear hypotheses of parameters when between-individuals design matrices are orthogonal.

Date received: October 9, 2000