A joint model for returns and realized measures of volatility: Considering dependencies among innovations

作者：Qu H^*; Li X D

来源：International Journal of Applied Mathematics and Statistics, 2013, 33(3): 1-7.

• Continuous-time volatility
• Dependency relationship
• Estimation performance
• Financial assets
• High frequency data
• Independent innovation
• Joint models
• Jump volatility
• Maximum likelihood estimation method
• Realized volatility
• Research models

Returns and volatilities of financial assets are critical considerations for investment and risk management. Most of the current research models returns and volatilities separately, implicitly assuming that their innovations are independent. We empirically investigate the dependency relationship among innovations of returns and realized measures of volatility using CSI-300 index 5-minute high-frequency data from 8 April 2005 to 23 May 2011. Based on the empirical results, we propose a joint model for returns and realized measures of volatility which characterizes the dependencies among innovations and provide the maximum likelihood estimation method. Empirical results confirm that the joint model effectively separates out the actual independent innovations and improves the parameter estimation performance.

• 出版日期2013
• 单位南京大学