模型平均方法及其在预测中的应用

统计研究 ›› 2011, Vol. 28 ›› Issue (6): 98-103.

模型平均方法及其在预测中的应用

张新雨邹国华

出版日期:2011-06-15 发布日期:2011-06-09

Model Averaging Method and Its Application in Forecast

Zhang Xinyu & Zou Guohua

Online:2011-06-15 Published:2011-06-09

摘要/Abstract

摘要： 内容提要：模型平均方法是当代统计学和计量经济学界研究的国际前沿问题，在经济、金融、生物、医学等领域有着广泛的应用前景。本文着重介绍几种常用的和最新的模型平均方法，并把它们应用于我国的粮食产量预测，取得了比较好的预测效果，表明模型平均方法为准确的预测分析提供了有力的工具，将对规避管理中可能出现的风险和偏差提供重要的技术支持。

关键词: 关键词：模型平均, 组合预测, 权重选择

Abstract: Abstract: Model averaging is a frontier topic in Statistics and Econometrics. It has wide applications in many fields like Economics, Finance, Biology and Medicine. This paper focuses on introducing some commonly used and newly proposed model averaging technologies, and applying them to forecast the grain output of China. The results on the forecasts show good precision, suggesting that the model averaging method is a powerful tool for practical forecast and provides an important technical support for avoiding the risk in management.

Key words: Key words: Model Averaging, Forecast by Combining, Weight Choice

张新雨邹国华. 模型平均方法及其在预测中的应用[J]. 统计研究, 2011, 28(6): 98-103.

Zhang Xinyu & Zou Guohua. Model Averaging Method and Its Application in Forecast[J]. , 2011, 28(6): 98-103.

[1]	钟波, 肖智. 一种基于粗糙集理论的组合预测方法[J]. 统计研究, 2002, 19(11): 37-39.
[2]	董景荣;杨秀苔. 基于T-S模糊模型的非线性组合预测方法研究[J]. 统计研究, 1999, 16(8): 55-57.
[3]	王明涛 . 确定组合预测权系数最优近似解的方法研究[J]. 统计研究, 1999, 16(7): 43-48.