Two-sided learning and short-run dynamics in a New Keynesian model of the economy

•We study asymmetric information and two-sided learning in a New Keynesian model.•Agents use econometric models to form beliefs about the unknown equilibrium dynamics.•In simulations, two-sided learning alters the short run dynamics of the model.•The impact is larger when policymakers' beliefs are still adjusting towards the SCE.

We investigate the role of asymmetric information and learning in a New Keynesian framework in which private agents and the central bank have imperfect knowledge of the economy. We assume that agents employ the data that they observe to form beliefs about the relationships that they do not know, use their beliefs to decide on actions, and revise these beliefs through a statistical learning algorithm as new information becomes available. Using simulations, we show that asymmetric information and learning can significantly change the dynamics of the variables of the model.