| Summary: | 碩士 === 國立高雄第一科技大學 === 財務管理所 === 90 === Abstract
This paper uses GARCH model to investigate the question of excessive implies persistence of volatility. Ninety one traded actively Taiwan stocks of SIMEX MSCI are considered and as already established in the literature, when volume traded is inserted in the GARCH(1,1) or EGARCH(1,1) model for returns, the estimated persistence is decreased since volume is affected also by within-the-day price movements. It is concluded that the difference between the opening price and the closing price of the previous day accounts also for most of the persistence in the autoregressive conditional hetero-skedasticity. The GARCH(1,1) or EGARCH(1,1) model is found to be more appropriate than traditional statistical models because it is capable of mimicking observed statistical characteristics of many time series of financial assets. The result of this study indicated that the change process of price and volume are best described by a GARCH(1,1) or EGARCH(1,1) model. Evidence is also exhibited that ONIt (over night index ) can help to explain the price volatility better than volume or other proxies.
|
|---|
