Forecasting GDP with the Leading Indicators: A VAR Approach

Abstract

The Conference Board’s Leading Economic Indicators Index suffers from construction flaws, which reduce its predictive power as well as one’s ability to interpret its signals. This paper develops a vector autoregression model to address these problems. The model’s out-of-sample GDP forecasts, using revised data, are found to outperform other private-sector forecasters on average over the period considered.