Adaptive learning and macroeconomic policy in DSGE models

The Global Financial Crisis of 2007-08 highlighted the shortcomings of standard macroeconomic models within the Rational Expectations framework, commonly used at central banks for forecasting and policy analysis. The goal of this thesis is to introduce alternative models of expectation formation mechanisms to improve the performance of operational macro models, and to give new insights into monetary and fiscal policy. A special emphasis is put on the estimation of adaptive learning and heterogeneous expectations models, with the aim of increasing their empirical relevance and popularity among central banks. The research in this thesis urges policymakers to expand their macroeconomic policy toolkit with alternative forms of bounded rationality to improve the robustness and resilience of macro models, in particular during times of turmoil and increased uncertainty.