Estimation of Stock Volatility Using Different Sample Frequency

• Main Authors: Wei-Chen Liao, 廖偉真
• Other Authors: Li-Fen Lei
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/23223206593063745695

碩士 === 國立臺灣大學 === 農業經濟學研究所 === 97 === Abstract The paper evaluates the performance of conditional variance models using high-frequency data of the TAIEX and attempts to determine the optimal sampling frequency for the best daily volatility forecast. In this paper, we use GARCH, T-GARCH and E-GARCH model to measure and also add volume and interaction term. Form the analysis, is found that sampling at 5 minutes gives the best forecast for volatility. Our analysis also suggests that volume and interaction term would give better estimates of volatility with lower forecast error estimate.