基于Levy-GARCH模型的上证50ETF市场跳跃行为与波动特征研究

上证ETF50期权的发行推动学界开始关注其标的资产的波动特点。本文引入带跳跃的Levy-GARCH非高斯条件异方差模型，结合Fourier数值极大似然估计及回溯测试，对上证50ETF的跳跃和波动特征进行实证分析，并与上证综指、深证成指进行比较。研究结果表明，与国内主要市场指数相比，上证50ETF市场同样存在显著的条件异方差效应和随机跳跃行为，但波动率并不存在显著的杠杆效应。本文通过与多个市场和行业指数进行对照比较，并从行业特征、成份股特征、市场机制特点等角度解释了上证50ETF杠杆效应不显著的原因。

Studies on Volatility Features and Jump Behavior of Shanghai 50ETF Market Based on Levy-GARCH Model

Volatility is one of the significant features of the capital market. It is important to characterize the volatility of financial markets accurately which plays an important role in effective risk management and rational derivatives pricing. ETF50 option listing begins to attract wide attention from academics to focus on the volatility features of underlying assets. Therefore, in order to study the random jump behavior and the volatility features of Shanghai 50ETF market, non-Gaussian Levy-GARCH model is introduced, such as Merton jump diffusion model, combined with maximum likelihood estimate with Fast Fourier and back testing, compared with Shanghai composite index and Shenzhen composite index, to analyze volatility features of Shanghai 50ETF market. Return series of several markets in China during the period of 2005 to 2016 are investigated. Research results show that fat tail, long memory, clustering effect, conditional heteroskedasticity and random jump behavior are well reflected in Shanghai 50ETF market, meanwhile, leverage effect does not exist. For exploring reasons why this phenomenon exists, the analysis will be presented in the following aspects:industry characteristics, features of constituent stocks and market mechanism. These researches will contribute to further development about the issue of derivatives pricing and risk management based on open-end funds.