Investigation of Intraday Volatilities of Oil Futures Markets in Taiwan

• Main Authors: LIN, SIAN-WEI, 林顯衛
• Other Authors: YANG, MING-JING
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/d9v52b

碩士 === 逢甲大學 === 財務金融學系 === 107 === Various volatility models are frequently used in the previous literature to predict the volatility of the spot markets. Recently, the volatility forecasting models have been gradually applied to the financial derivatives markets. The intraday prices of the futures markets are changing rapidly, and the intraday volatility of the futures markets may have more information than the daily volatility. Brent crude oil futures began trading on the Taiwan Futures Exchange in July 2018. Very little literature has discussed the volatility of the Taiwan’s crude oil futures markets. Therefore, this study focuses on the investigation of the intraday volatilities of the crude oil futures markets in Taiwan. This research is based on the HAR-RV models proposed by Corsi (2009). Then, the put-call ratios formed by the trading volume and open interest of the at-the-money nearby contracts of Taiwan weighted index options and the Baker and Wurgler (2007) index of sentiment levels are also included separately in the models as the investors’ sentiment variables, to examine the influence on volatility from different investors’ sentiment levels. Besides, the CBOE crude oil ETF volatility index, the trading volume, open interest, and daily returns of Taiwan's crude oil futures markets are also included in the models to explore the intraday volatilities of the Brent crude oil futures markets in Taiwan. The empirical results of the study show that the characteristics of volatility clustering and long memory are presented in the volatility of Taiwan’s Brent crude oil futures markets. After the CBOE crude oil ETF volatility index is added in the models, the explanatory power of the models for the realized volatility has been greatly improved. In addition, after the different levels of the investors’ sentiment variables are incorporated into the models, the adjusted R^2 of the models with the investors’ sentiment variables is higher than that of the models without the variables. Therefore, it indicates that the investors’ sentiment variables can also further enhance the explanatory ability of the models.