ABSTRACT
Chapters 3 to 7 have looked at ways to model and analyze different types of multivariable data in which there is a single response variable and a number of explanatory variables, and only the response is considered a random variable. Chapters 9 to 11 look at ways to explore and uncover possible structure in multivariate data in which there may be many variables but no division into response and explanatory variables, and all the variables are random variables. In this chapter we look at ways of modeling a type of data that comes somewhere between the multivariable and the multivariate. As with the former, there is a division into explanatory and response variables, but the response variable (and possibly some of the exploratory variables) is observed more than once on each individual in the study; consequently, the response is multivariate.
