Inferring monetary policy objectives with a partially observed state

Accounting for the uncertainty in real-time perceptions of the state of the economy is believed to be critical for monetary policy analysis. We investigate this claim through the lens of a New Keynesian model with optimal discretionary policy and partial information. Structural parameters are estimated using a data set that includes real-time and ex post revised observations spanning 1965–2010. In comparison to a standard complete information model, our estimates reveal that under partial information: (i) the Federal Reserve demonstrates a significant concern for stabilizing the output gap after 1979, (ii) the model׳s fit with revised data improves, and (iii) the tension between optimal and observed policy is smaller.