We develop coincident and leading employment indexes for the Connecticut economy. Four employment-related variables enter the coincident index while five employment-related variables enter the leading index. The peaks and troughs in the leading index lead the peaks and troughs in the coincident index by an average of 3 and 9 months. Finally, we use the leading index in vector-autoregressive (VAR) and Bayesian vector-autoregressive (BVAR) models to forecast the coincident index, nonfarm employment, and the unemployment rate.
Dua, Pami and Miller, Stephen M., "Forecasting and Analyzing Economic Activity with Coincident and Leading Indexes: The Case of Connecticut" (1995). Economics Working Papers. 199505.