Adaptive Learning, Model Uncertainty and Monetary Policy Inertia in a Large Information Environment, mimeo

Comments are Welcome!) This paper studies monetary policy in a large information environ-ment. We allow the central bank to exploit information coming from several indicator variables. Model uncertainty is extremely pervasive in this large information world. We take it into account, by estimating the whole set of possible models through Bayesian Model Averaging, implemented with a Markov Chain Monte Carlo Model Composition (MC3) strategy. The central bank can learn parameters and models through time. In this setting, we aim at evaluating if the introduction of a large information set can, at least partially, explain the optimality of observed monetary policy inertia