ABSTRACT
Multivariate data arise when researchers measure several variables on each individual in their sample. This would appear to imply that in previous chapters of this book we have already been concerned with multivariate data. But this is not the case; previously, the data sets have involved a single response variable and a set of explanatory variables or covariates and, as explained in several places, it is strictly only the response that is considered to be a random variable. With multivariate data,
all
the measurements are regarded as random variables, with a particular multivariate probability distribution, for example, the
multivariate normal
(see Everitt, 2001)
.
