Estimating the Output Gap Using Business Survey Data - A Bivariate Structural Time Series Model for the German Economy

This paper deals with the estimation of the output gap. We use uni- and bivariate unobserved components models in order to decompose the observed German GDP-series into trend, cycle and seasonal components. The results show that using the ifo business assessment variable as an indicator for the cycle the estimation of the output gap is much more precise and out-of-sample forecasts exhibit smaller prediction errors.