利用遗传算法对上证综指EGARCH模型的实证分析  

An empirical study of the EGARCH model Shanghai stock composite index by the improved real-coded genetic algorithm

作　　者：汪飞星[1] 李勇静[1] 

机构地区：[1]北京科技大学应用科学学院数学力学系,北京100083

出　　处：《辽宁师范大学学报（自然科学版）》2007年第1期28-30,共3页Journal of Liaoning Normal University:Natural Science Edition

摘　　要：比较了改进的遗传算法和BHHH算法对EGARCH模型的估计效果.结果显示,改进的遗传算法优于BH-HH算法.然后分别用在正态分布t、分布和广义误差分布(GED分布)假设下的EGARCH模型对上证综指进行实证分析.分析结果表明,EGARCH模型在厚尾分布假设下对序列的拟合效果比在正态分布假设下的效果好.上证综指收益率序列具有明显的杠杆效应.This paper estimates the parameters of EGARCH-t model by the improved genetic algorithm and the BHHH algorithm,then the results are compared. It shows that the former performs better than the later. Then the series of the return rates of Shanghai stock index is analyzed by EGARCH-n. EGARCH-t and EGARCH-GED models, respectively. The results show that EGARCH model with fat-tailed distribution is better than those with normal distribution, and the series of the return rates of Shanghai stock composite index has obvious leverage effect.

关 键 词：遗传算法 T分布 广义误差分布 EGARCH模型 

分 类 号：O211.3[理学—概率论与数理统计]