Inflation dynamics and adaptive expectations in an estimated DSGE model

We estimate a "hybrid expectations" version of the Smets and Wouters (2007) model in which a subset of agents employ simple moving-average forecast rules that place a significant weight on the most recent data observation. We show that the overall fit is improved relative to an otherwise similar version in which all agents have fully rational expectations. In-sample and out of-sample analyses show the superiority of the hybrid expectations model in generating an expected inflation series that more closely tracks expected inflation from the Survey of Professional Forecasters.