主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2011, Vol. 19 ›› Issue (3): 11-18.

• 论文 • 上一篇    下一篇

基于ICA模型的国际股指期货及股票市场对我国股市波动溢出研究

柴尚蕾, 郭崇慧, 苏木亚   

  1. 大连理工大学系统工程研究所, 辽宁大连116024
  • 收稿日期:2010-10-11 修回日期:2011-04-12 出版日期:2011-06-30 发布日期:2011-06-30
  • 作者简介:柴尚蕾(1982- ),女(汉族),山东济南人,大连理工大学系统工程研究所博士生,研究方向:时间序列分析、金融数据挖掘.
  • 基金资助:

    国家自然科学基金资助项目(70871015);中央高校基本科研业务费专项资金资助(DUT11SX04)

Volatility Spillover from International Stock Index Futures and Spot Markets to Chinese Stock Market Based on ICA Model

CHAI Shang-lei, GUO Chong-hui, SU Mu-ya   

  1. Institute of Systems Engineering, Dalian University of Technology, Dalian 116024, China
  • Received:2010-10-11 Revised:2011-04-12 Online:2011-06-30 Published:2011-06-30

摘要: 将独立成分分析(ICA)方法引入金融衍生品市场与基础市场之间的波动溢出研究,克服了传统方法解决高维金融时间序列波动问题时的障碍。通过与VECH、BEKK和DCC等传统多元GARCH模型的对比分析,本文所建立的ICA-EGARCH-M模型在解决高维问题时体现出一定的优势。在实证研究中,应用该模型考察了美国、英国、日本和中国香港的股指期货市场及其股票市场对我国股票市场的共同波动溢出。结果表明ICA-EGARCH-M模型不仅验证了波动溢出效应的存在,而且反映出了波动溢出的主要来源,能够较好地解决高维金融时间序列数据的波动溢出问题。

关键词: 金融市场, 股指期货, 波动溢出, 独立成分分析, GARCH模型

Abstract: Independent Component Analysis(ICA)is introduced to study volatility spillovers from financial derivative markets to basic markets.It remedies the deficiency of using traditional methods to solve high dimensional financial time series volatility problem in the past.By comparing with multiv ariate GARCH models,such as VECH,BEKK and DCC,ICA-EGARCH-M model in this paper shows some advantages of solving high dimensional problem.In empirical study,ICA-EGARCH-M model is employed to examine volatility spil lovers effects from stock index futures and spot markets of the US,UK,Japan and Hongkong to Chinese stock market.The results show that the ICA-EGARCH-M model not only confirms that there exists volatility spillovers,but also reflects the main resource of volatility spillovers.It can better resolve volatility spillovers problem of high dimensional financial time series.

Key words: financial markets, stock index futures, volatility spillovers, independent component analysis, GARCH model

中图分类号: