ABSTRACT
In Chapter 8 we discussed the methodology for analysing experiments with a p-variate response, the methodology being based on multivariate normal-distribution theory. In this chapter we will be concerned with further methodology also based on normal theory but with the emphasis on the conditional distribution of a subset of the response variates given the values of the others. The techniques to be discussed are multivariate regression, multivariate analysis of covariance (MANOCOVA) and the test for additional information. The approach will again be heuristic with the procedures being presented as generalizations of univariate procedures.
