Adaptive Inflation Expectations Under Structural Changes: The Case of Finland

Abstract

Via the use of rolling regression technique and a specific procedure for analyzing strong structural breaks in a univariate time series model, we forecast the rate of future inflation in Finland for the time period of unregulated financial markets since the beginning of 1987. We are able to label the identified structural changes in the data generating process (DGP) of inflation both with economic events and changes in the main leading inflation indicators. In terms of forecast performance, the role of structural breaks in the coefficient variation of an ARIMA model for inflation seems to be stronger in the shorter than one-year forecasting horizon. However, the estimated final intervention model does not in all cases yield significantly better forecasts than pure rolling regression technique without identification of the strong breaks. When treating the obtained forecasts as proxies for adaptive inflation expectations, in the comparison to the actual future inflation rate, the generated expectation series perform better than some non-continuous series of inflation expectations based on questionnaires, i.e., inflation expectation surveys.