BAYESIAN REDUCED RANK REGRESSION IN ECONOMETRICS

The reduced rank regression model arises repeatedly in theoretical and applied econometrics. To date the only general treatments of this mod el have been frequentist. This paper develops general methods for Baye sian inference with noninformative reference priors in this model, bas ed on a Markov chain sampling algorithm, and procedures for obtaining predictive odds ratios for regression models with different ranks. The se methods are used to obtain evidence on the number of factors in a c apital asset pricing model.