Combination of Forecasts using Bayesian Model Averaging (original) (raw)
Related papers
To combine or not to combine: selecting among forecasts and their combinations
International Journal of Forecasting, 2005
Much research shows that combining forecasts improves accuracy relative to individual forecasts. In this paper we present experiments, using the 3003 series of the M3-competition, that challenge this belief: on average across the series, the best individual forecasts, based on post-sample performance, perform as well as the best combinations. However, this finding lacks practical value since it requires that we identify the best individual forecast or combination using post sample data. So we propose a simple model-selection criterion to select among forecasts, and we show that, using this criterion, the accuracy of the selected combinations is significantly better and less variable than that of the selected individual forecasts. These results indicate that the advantage of combining forecasts is not that the best possible combinations perform better than the best possible individual forecasts, but that it is less risky in practice to combine forecasts than to select an individual forecasting method.
Combination of forecasts: A bibliography
Research Papers in Economics, 1998
During the past thirty years, there has been considerable concern about combination of forecasts. Many of the articles and books dedicated to this specific area explain and demonstrate that combining multiple individual forecasts can improve forecast accuracy. The improvement in accuracy mainly depends on forecast combination techniques which range from simple combinations like averaging the forecasts to more complex ones that use the Bayesian approach. This paper provides a bibliography of selected articles and books related to the combination of forecasts in various disciplines and is intended to be a catalog for locating contributions in research areas focusing on the theory and applications of combining forecasts. The bibliography includes recent articles and is as up-to-date as possible.
Does Forecast Combination Improve Norges Bank Inflation Forecasts?*
Oxford Bulletin of Economics and Statistics, 2012
We develop a system that provides model-based forecasts for inflation in Norway. We recursively evaluate quasi out-of-sample forecasts from a large suite of models from 1999 to 2009. The performance of the models are then used to derive quasi real time weights that are used to combine the forecasts. Our results indicate that a combination forecast improves upon the point forecasts from individual models. Furthermore, a combination forecast out-performs Norges Bank's own point forecast for inflation. The beneficial results are obtained using a trimmed weighted average. Some degree of trimming is required for the combination forecasts to out-perform the judgmental forecasts from the policymaker.
To combine or not to combine? issues of combining forecasts
Journal of Forecasting, 1992
ABSTRACT This paper addresses issues such as: Does it always pay to combine individual forecasts of a variable? Should one combine an unbiased forecast with one that is heavily biased? Should one use optimal weights as suggested by Bates and Granger over twenty ...
This paper proposes a strategy to increase the efficiency of forecast combining methods. Given the availability of a wide range of forecasting models for the same variable of interest, our goal is to apply combining methods to a restricted set of models. To this aim, an algorithm procedure based on a widely used encompassing test (Harvey, Leybourne, Newbold, 1998) is developed. First, forecasting models are ranked according to a measure of predictive accuracy (RMSFE) and, in a consecutive step, each prediction is chosen for combining only if it is not encompassed by the competing models. To assess the robustness of this procedure, an empirical application to Italian monthly industrial production using ISAE short-term forecasting models is provided.
Combining Forecasting Procedures: Some Theoretical Results
2000
We study some methods of combining procedures for forecasting a continuous random variable. Statistical risk bounds under the square error loss are obtained under mild distributional assumptions on the future given the current outside information and the past observations. The risk bounds show that the combined forecast automatically achieves the best performance among the candidate procedures up to a constant factor and an additive penalty term. In term of the rate of convergence, the combined forecast performs as well as if one knew which candidate forecasting procedure is the best in advance.