Learning in Survey Expectations of Inflation
|Speaker:||Dr Krisztina Molnar, NHH Norwegian School of Economics and Business Administration|
|Date:||Friday 19 March 2010|
|Location:||Xfi Lecture Theatre|
We examine how agents learn in different environments by calibrating adaptive learning algorithms using survey expectations of inflation. Our result is that compared to developed economies, in Transition and Latin American economies with high inflationary episodes private agents pay more attention to recent data when they form their inflationary expectations. In particular, we find that the higher is the variance of inflation the more private agents discount past data. In a small model with permanent and transitory shocks we show that this relationship is a result of two effects: (1) the more important are permanent shocks compared to transitory shocks the bigger is the optimal gain parameter and the bigger is the variance of inflation, (2) when expectations use a bigger gain parameter this amplifies the effect of shocks on the variance of inflation expectations and inflation.