Tests of Rank in Reduced Rank Regression Models

Recently, there has been renewed research interest in the development of tests of the rank of a matrix based on a root- Tconsistent estimator. This paper evaluates the performance of some asymptotic tests of rank determination in reduced rank regression models through simulation experiments together with their bootstrapped versions. The bootstrapped procedures significantly improve upon the performance of the corresponding asymptotic tests. The tests of rank considered are applied to construct reduced rank VAR models of leading indicators of UK economic activity and these more parsimonious multivariate representations improve the forecasting performance of VAR models.