Research on Volatility Modeling of China's Oil Industry

Abstract

Stock prices in the oil industry tend to fluctuate frequently, but violent price fluctuations will also have a serious impact on the lives of investors and residents. Companies in the oil industry often show many similar or related characteristics due to wilderness factors. This correlation between companies includes a risk spillover effect. This spillover effect should play a vital role in risk management and investment decision-making. If an econometric model to measure this effect can be got, the existing fundamental analysis will produce the effect of qualitative change: from qualitative analysis to quantitative analysis. Firstly, this paper will verify that this spillover effect is statistically real through the Granger causality test. Next, it needs to quantify the risk of a single company through the existing volatility models, including GARCH family models applicable for low-frequency data and HAR models applicable for high-frequency data. It is found that there is a causal relationship between the volatility of oil enterprises in China's stock market. In addition, through the volatility modeling of Sinopec, it is found that its logarithmic return series has obvious leverage effect, and its logarithmic return distribution also has obvious characteristics of high tail thick peak. Finally, The HAR model is used to predict the realized volatility series of Sinopec, and the final out of sample prediction RMSE is 4.49e-05.