Key points are not available for this paper at this time.
SUMMARY A method of analysis of 'growth curves' that utilizes the technique of analysis of covariance is developed and illustrated. The method yields results identical to those obtained by weighting inversely by the sample covariance matrix, but has the additional feature of allowing flexibility in weighting by choosing subsets of covariates that have special properties. Thus, exploitation of Rao's observation that selected subsets of the covariates may in some cases yield 'better' estimates than the complete set is easily implemented. Examples are included to illustrate the calculations and how to determine when it will be advantageous to use all the vectors in the error space as covariates (weight inversely by the sample covariance matrix) or to use only a subset, and how to choose the 'best' subset from examination of sample covariance or correlation matrices.
Grizzle et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: