Combining Economic Forecasts
Combining Economic Forecasts
复制标题
DOI：
10.1080/07350015.1986.10509492
复制
发表时间：
1986
影响因子：
3
通讯作者：
R. Clemen;R. L. Winkler
中科院分区：
文献类型：
--
作者：
R. Clemen;R. L. Winkler
结合预测的一种方法可能会或可能不会说明预测之间的依赖性和精度不同。在本文中，我们在结合四个主要计量经济学模型的GNP预测的背景下测试了各种此类方法。这些方法包括一种预测误差共同正态分布，该模型的几种变体以及一些更简单的过程以及基于预报员交换性的先验分布的贝叶斯方法。结果表明，一个简单的平均值，具有独立性假设的正常模型，并且贝叶斯模型的性能优于此处研究的其他方法。
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.