Combining Economic Forecasts
- 1 January 1986
- journal article
- research article
- Published by Taylor & Francis in Journal of Business & Economic Statistics
- Vol. 4 (1) , 39-46
- https://doi.org/10.1080/07350015.1986.10509492
Abstract
A method for combining forecasts may or may not account for dependence and differing precision among forecasts. In this article we test a variety of such methods in the context of combining forecasts of GNP from four major econometric models. The methods include one in which forecasting errors are jointly normally distributed and several variants of this model as well as some simpler procedures and a Bayesian approach with a prior distribution based on exchangeability of forecasters. The results indicate that a simple average, the normal model with an independence assumption, and the Bayesian model perform better than the other approaches that are studied here.Keywords
This publication has 10 references indexed in Scilit:
- Improved methods of combining forecastsJournal of Forecasting, 1984
- Averages of Forecasts: Some Empirical ResultsManagement Science, 1983
- Optimal Aggregation of Money Supply Forecasts: Accuracy, Profitability and Market EfficiencyThe Journal of Finance, 1983
- The Combination of ForecastsJournal of the Royal Statistical Society. Series A (General), 1983
- Combining Probability Distributions from Dependent Information SourcesManagement Science, 1981
- Combining Expert Judgments: A Bayesian ApproachManagement Science, 1977
- Experience with Forecasting Univariate Time Series and the Combination of ForecastsJournal of the Royal Statistical Society. Series A (General), 1974
- Bayes Estimates for the Linear ModelJournal of the Royal Statistical Society Series B: Statistical Methodology, 1972
- A Bayes Approach for Combining Correlated EstimatesJournal of the American Statistical Association, 1965
- Almost Linearly-Optimum Combination of Unbiased EstimatesJournal of the American Statistical Association, 1961