Information, data dimension and factor structure

This paper employs concepts from information theory to choosing the dimension of a data set. We calculate relative measures of information in the data in terms of eigenvalues and derive criteria to determine the `optimal' size of the data set, in particular whether an extra variable adds information. The methods can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a macroeconomic data set on The Netherlands.