A composite approach to forecasting state government revenues: Case study of the Idaho sales tax

Fiscal problems have led to increased reliance on economic and revenue forecasting by state governments in recent years. As a means of improving accuracy, many forecasters use alternative outlooks. Composite modeling goes a step further and allows analysts to systematically combine two or more forecasts. This paper examines the effectiveness of composite forecasting of sales tax revenues in Idaho. Base line projections are provided by an econometric model and a univariate time series model. The composite forecasts are found to outperform both base line forecasts. The combined forecasts are also found to be more accurate than the executive branch forecasts actually utilized from 1982 through 1985.