Temporal Aggregation of GARCH Process: An Empirical Study of the Stock Markets in Greater China

• Main Authors: Ching-Lian Lai, 賴警聯
• Other Authors: Chung-Chu Chuang
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/09850273815529921880

碩士 === 真理大學 === 管理科學研究所 === 93 === This study is to investigate volatility-clustering and the parameters consistency between implied estimates on the temporal aggregation and direct estimates of GARCH(1,1) models at four frequencies(daily, weekly, bi-weekly and monthly) written on Greater China Stock markets. The resultant shows the data generating process of the daily stock indices of Greater China are close to IGARCH model. There are not consistent between the direct estimates and the implied estimates based on the daily frequency data other then Shanghai A stock index. In addition, the parameters sum of conditional heteroscedasticiity function tend to zero as aggregation level m approach to infinity, and conditional heteroscedasticity almost disappears.