37 Pages Posted: 28 May 2011 Last revised: 13 Apr 2013
Date Written: April 7, 2011
We assess the forecast ability of Norges Bank’s regional survey for inflation, GDP growth and the unemployment rate in Norway. We propose several factor models based on regional and sectoral information given by the survey. The analysis identifies which information extracted from the ten sectors and the seven regions performs particularly well at forecasting different variables and horizons. Results show that several factor models beat an auto-regressive benchmark in forecasting inflation and unemployment rate. However, the factor models are most successful in forecasting GDP growth. Forecast combinations based on past performances give, in most cases, more accurate forecasts than the benchmark, but they never give the most accurate forecasts.
Keywords: Factor models, macroeconomic forecasting, qualitative survey data
JEL Classification: C53, C80
Suggested Citation: Suggested Citation
Wulfsberg, Fredrik and Ravazzolo, Francesco and Martinsen, Kjetil, Forecasting Macroeconomic Variables Using Disaggregate Survey Data (April 7, 2011). Norges Bank Working Paper 2011/04 . Available at SSRN: https://ssrn.com/abstract=1847595 or http://dx.doi.org/10.2139/ssrn.1847595