A study the Expiration-Day Effect of Taiwan Index Futures when Settlement Rule Changes: Use The GARCH Model

• Main Authors: Yu-Hong Wu, 吳昱宏
• Other Authors: En-Der Su
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/86002461028121106767

碩士 === 國立高雄第一科技大學 === 風險管理與保險研究所 === 99 === This paper studies the effect of expiration day of Taiwan Stock Index Futures (TX), the pre-period and the post-period of settlement rule changes at 2008.11.21. In particular, The earlier studies mostly examined significant differences in the statistical properties of asset returns (for instance, mean and variance) during expiration and non-expiration days, but there are some flaws when modelling the data generation process. The data of futures fit the traits of finance time series statistical data, so we propose to address the shortcoming by examining the expiration day effect from a GARCH (Generalized Auto Regressive Conditional Heteroskedastic) framework. We use both daily and high frequency (5 min) data on TX. The main evidences are: 1. In return: The expiration-day effects did not have any significant positive or III negative effect in daily and per 5min data on the pre-period and the post-period of settlement rule changes. 2. In the volatility: The expiration-day effects have significant downward pressure in daily data on the pre-period and the post-period of settlement rule changes; The expiration-day effects did not have significant pressure in high frequency(5 min) data on the pre-period of settlement rule changes, but it has significant downward pressure on the post-period of settlement rule changes. Based on this findings, the volatility is significant lower after settlement rule changes, this explained that the new settlement rule is advantageous in reduces the expiration-day effect.