股票市场资产收益的跳跃行为研究

发布时间：2018-03-14 06:13

本文选题：资产收益　切入点：跳跃行为　出处：《经济研究》2010年04期 　论文类型：期刊论文

【摘要】：本文通过构建混合GARCH跳跃模型和实证研究来探讨股票市场资产收益的跳跃行为。本文构建了一种新型的混合GARCH跳跃模型,即ARVI-GARCH-Jump模型。模型中,以依赖于条件波动率的自回归条件跳跃强度过程控制跳跃行为的发生概率,以依赖于跳跃行为、带位移特征的GARCH过程控制正常漂移扰动的条件方差。模型综合考虑跳跃行为的时变特征、集聚效应和条件波动率的非对称特征、位移特征。作为对现有GARCH跳跃模型的主要扩展,模型考虑条件波动率与跳跃行为之间相互的直接回馈效应,即条件波动率对跳跃行为预期的回馈以及跳跃行为对波动率预期的回馈。与此同时,本文采用上证综合指数、香港恒生指数、台湾加权指数、道琼斯工业指数和纳斯达克综合指数的数据进行了实证检验。实证结果表明,条件波动率与跳跃行为之间相互的直接回馈效应得到证实。此外,跳跃行为存在时变特征和集聚效应,条件波动率存在传统的非对称效应和位移效应,并且非对称效应的程度因跳跃行为的发生而加剧或者减缓。
[Abstract]:This paper discusses the jump behavior of asset returns in stock market by constructing a hybrid GARCH jump model and empirical research. In this paper, a new hybrid GARCH jump model, ARVI-GARCH-Jump model, is constructed. The probability of jumping behavior is controlled by the autoregressive conditional jump intensity process, which depends on the conditional volatility, and depends on the jump behavior. The GARCH process with displacement characteristics controls the conditional variance of normal drift disturbance. The model takes into account the time-varying characteristics of jumping behavior, the asymmetric characteristic of agglomeration effect and conditional volatility, and the displacement characteristic. It is the main extension of the existing GARCH jump model. The model considers the direct feedback effect between conditional volatility and jump behavior, that is, the feedback of conditional volatility on hopping behavior and the feedback of jumping behavior on volatility expectation. At the same time, the Shanghai Composite Index is used in this paper. The data from Hong Kong Hang Seng Index, Taiwan weighted Index, Dow Jones Industrial Index and Nasdaq Composite Index are empirically tested. The empirical results show that the direct feedback effect between conditional volatility and jumping behavior is confirmed. Jumping behavior has time-varying characteristics and agglomeration effect, conditional volatility has traditional asymmetric effect and displacement effect, and the degree of asymmetric effect is aggravated or slowed down by the occurrence of jump behavior.
【作者单位】：中国立信风险管理研究院;中山大学经济研究所;中山大学岭南学院;华南理工大学经济与贸易学院;金融工程研究中心;
【基金】：国家自然科学基金项目(70673116) 国家社会科学基金项目(09CJY013,08ATL007) 北京大学汇丰金融研究院2009年课题教育部人文社会科学研究项目(08JC790038,05JJD790075) 中山大学“985工程”产业与区域发展研究创新基地广东省哲学社会科学“十一五”规划青年项目(08YE-02) 广东高校优秀青年创新人才培育项目(WYM08084) 广东省自然科学基金项目(9151027501000032) 广东省普通高校人文社会科学重点研究基地经费资助成果之一
【分类号】：F224;F830.91

【相似文献】