Intraday Value at Risk using high frequency data with application to the Taiwan Stock Exchange

• Main Authors: Yi-Ting Yeh, 葉奕廷
• Other Authors: Hsin-Hue Chang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/72449229179398551099

碩士 === 銘傳大學 === 財務金融學系碩士班 === 99 === This study employs the log-WACD-ARMA-EGARCH and Monte Carlo method to calculate the intraday value at risk. The Intraday Value at Risk employing the high frequency data to calculate the risk, is generally used to measure the relevant risk. The advantage of the Monte Carlo method used in this study is that it can be applied to the nonlinear and nonnormal financial dataset. This enables simultaneously overcoming the problem of fat-tail and extreme conditions. We can simulate different price scenarios and calculate the intraday risk, solving the measurement of specific risk unique to the investors and market makers. There are twenty-four stocks of Taiwan Stock Exchange, this study pair-wisely compares the results of the intraday risk for the small and large capitalization, high and low turnover rate stocks. The empirical results indicate that Monte Carlo simulation intraday VaR is higher than GARCH method and historical simulation method. In different level of confidence, intraday VaR of large-capitalization stocks is higher than intraday VaR of small-capitalization stocks; intraday VaR of cold stocks is lower than intraday VaR of hot stocks; intraday VaR of small-capitalization and cold stocks is lower than intraday VaR of large-capitalization and hot stocks. This paper also found that GARCH method to estimate is more accuracy than others using backtesting methods.