Probit Based Time Series Models in Recession Forecasting - A Survey with an Empirical Illustration for Finland

Abstract

<p>This article surveys both earlier and recent research on recession forecasting with probit based<br />time series models. Most studies use either a static probit model or its extensions in order to<br />estimate the recession probabilities, while others use models based on a latent variable approach<br />to account for nonlinearities. Many studies find that the term spread (i.e, the difference<br />between long-term and short-term yields) is a useful predictor for recessions, but some recent<br />studies also find that the ability of spread to predict recessions in the Euro Area has diminished<br />over the years. Confidence indicators and financial variables such as stock returns seem to<br />provide additional predictive power over the term spread. More sophisticated models outperform<br />the basic static probit model in various studies. An empirical analysis made for Finland<br />strengthens the findings of earlier studies. Consumer confidence is especially useful predictor<br />of Finnish business cycle and the accuracy of the static single-predictor model can be improved<br />by using multiple predictors and by allowing the dynamic extension.<br /></p>