基于AR-EGARCH的空气温度预测模型  

The Air Temperature Forecasting Model Based on AR-EGARCH

作　　者：崔海蓉[1] 张京波[2] 何建敏[3] 

机构地区：[1]南京信息工程大学经济管理学院,江苏南京210044 [2]南京信息工程大学大气物理学院,江苏南京210044 [3]东南大学经济管理学院,江苏南京210096

出　　处：《统计与信息论坛》2013年第10期36-41,共6页Journal of Statistics and Information

基　　金：江苏省教育厅高校哲学社会科学基金项目<农业气象巨灾风险主体行为及管理机制设计--以江苏省为例>(2012SJB630047);国家自然科学基金项目<基于复杂网络的银行间传染风险及其演化模型研究>(71071034)

摘　　要：天气衍生产品定价极其复杂,其中与温度相关的产品是目前研究热点,其定价核心在于温度变量的精确预测。传统AR-GARCH温度预测模型难以描述温度变量波动率的非对称性。基于此,构建了AR-EGARCH温度预测模型,它能够描述波动率的非对称性,更好地反映温度变化过程。对中国东部南北线六个较发达的城市进行实证研究,结果表明:六个城市的温度变化具有明显的非对称性,AR-EGARCH模型无论是拟合还是预测效果都较传统的AR-GARCH模型更优。Weather derivatives pricing is extremely complex. Temperature-related products are the current research focus, and its pricing lies in the accurate prediction of the temperature variables. The traditional AR-GARCH temperature prediction model is difficult to describe the volatility asymmetry. Based on that, the AR-EGARCH temperature prediction model is constructed. It can describe the volatility asymmetry of the temperature variables and better reflects the temperature process. Empirical result based on the six developed cities of the north-south line in China＇s eastern shows that： the temperature change of the six cities has a significant asymmetry; whether it is fitting or prediction, AR- EGARCH model is better than the traditional AR-GARCH model.

关 键 词：天气衍生品 气温预测 波动率 非对称性 EGARCH 

分 类 号：F830[经济管理—金融学]