Finding good predictors for inflation by shotgun stochastic search∗,†

This paper evaluates a novel sampling algorithm, called shotgun stochastic search (S3), for Bayesian model averaging in the context of finding predictors for inflation when the set of potential predictors is large. This is a relevant case in the forecasting literature, where often hundreds of predictors are compared with autoregressive distributed lag models for inflation. With such a large model space, standard Bayesian approaches like MCMC model composition (MC3) tend to converge slowly. On the other hand, S3 systematically searches in the neighborhood of good models and concentrates on regions of high posterior probability in the model space. We carry out a Monte Carlo simulations to compare the computational effi ciency of S3 to MC3, based on standard data generating processes from the literature. When many potential predictors are available, S3 outperforms MC3. In an empirical exercise, we apply the two algorithms to find predictors for US inflation from a set of about one hundred indicators and their lags. S3 absorbs posterior mass much quicker than MC3 and makes Bayesian estimation of the standard inflation equations with many predictors computationally feasible.