Using Large Data Sets to Forecast Sectoral Employment

We use several models using Bayesian and classical methods to forecast employment for eight sectors of the US economy. In addition to standard vector-autoregressive and Bayesian vector autoregressive models, we also include the information content of 143 additional monthly series in some models. Several approaches exist for incorporating information from a large number of series. We consider two multivariate approaches – extracting common factors (principle components) and Bayesian shrinkage. After extracting the common factors, we use Bayesian factor-augmented vector autoregressive and vector error-correction models, as well as Bayesian shrinkage in a large-scale Bayesian vector autoregressive models. Using the period of January 1972 to December 1989 as the in-sample period and January 1990 to March 2009 as the out-of-sample horizon, we compare the forecast performance of the alternative models. Finally, we forecast out-of sample from April 2009 through March 2010, using the best forecasting model for each employment series as well as combined forecasts. We find that factor augmented models, especially error-correction versions, generally prove the best in out-of-sample forecast performance, implying that in addition to macroeconomic variables, incorporating long-run relationships along with short-run dynamics play an important role in forecasting employment. Forecast combination models, however, based on the simple average forecasts of the various models used, outperform the best performing individual models for six of the eight sectoral employment series.