Evaluating Automatic Model Selection

We evaluate automatically selecting the relevant variables in an econometric model from a large candidate set.  General-to-specific selection is outlined for a constant model in orthogonal variables, where only one decision is required to select, irrespective of the number of regressors (N < T) where T is the sample size, then evaluated in simulation experiments for N = 1000.  Comparisons with Autometrics (Doornik, 2009) show similar properties, but not restricted to orthogonal cases.  Monte Carlo experiments examine the roles of post-selection bias corrections and diagnostic testing, and evaluate Autometrics' capability in dynamic models by its cost of search versus costs of inference.