Modeling The Driving Mechanism Of International Oil Price And Its Impact On Stock Market Return

Crude oil is an important non-renewable energy source and one of the most important inputs for modern industrial production.Due to its scarcity and extreme importance to the real economy,it is often referred to as “black blood” or “black gold”.Due to the extreme importance of crude oil to the global real economy,fluctuations in crude oil prices have a huge impact on global macroeconomic and financial markets.China is a major oil importer,and its dependence on foreign oil has increased year by year.The uncertainty of future international oil prices will also have an increasingly deep impact on Chinese macroeconomic and financial markets.What is more important is that the simultaneous possession of commodity,political and financial properties of crude oil determines that crude oil prices are affected by fundamental factors,but also sensitive to the combined effects of geopolitical events,speculative positions and fluctuations in the US dollar exchange rate.The uncertainty of oil prices is increasing.In this context,it is important to answer the following questions: What is the oil price driving mechanism? How to better predict oil price volatility? What is the impact of crude oil price shocks on stock market returns? This paper discusses these issues,and the research content mainly includes three parts.The first part is the study of crude oil price volatility mechanism,including significant drivers of oil price volatility,the impact of environment on crude oil market and investor concerns about the impact of crude oil volatility.The second part focuses on the impact of regime switching on oil price volatility forecasting.The third part studies the impact of crude oil price shocks on the stock market.Specifically,the second chapter studies the influencing mechanism of oil price bubbles on oil price fluctuation.This paper proposes the log-periodic power law(LPPL)model based on state space model.This model can effectively describe the evolution process of potential bubbles while dynamically interpreting the oil price driving mechanism.Predict the point at which the potential bubble breaks.Firstly,by introducing the state space model into the D test method,the super-exponential growth of the oil price series is dynamically measured to exclude some sample intervals without foam,and then the envelope function analysis is used to detect the logarithmic periodicity of the oil price sequence so that we can further determine the existence of oil price bubbles.Finally,the LPPL model characterizes the dynamic evolution of oil price bubbles and predicts the breakpoint of potential bubbles.The log-periodic power-law model based on the state space model proposed in this paper provides a new tool for the oil price bubble detection model.This part of the research helps oil market investors and regulators to better protect against extreme risks.The third chapter investigates the impact of the environment in the location of crude oil futures trading on oil prices.This paper selects the Air Quality Index(AQI)and the Seasonal Affective Disorder(SAD)index as the proxy of the environment,and uses the Z test,linear regression and quantile regression model to study the impact of the environment on the crude oil market returns and volatility.Meanwhile,this paper compares the response of Chinese crude oil market to environmental impacts with that of American crude oil market.The results show that the environment has a significant impact on the crude oil market returns and volatility,and the role of the environment on Chinese and American crude oil market is not completely consistent.This results of this study impact the traditional effective market hypothesis and help to further reveal the price-driving mechanism of the crude oil market.The fourth chapter examines the impact of investor concerns on crude oil prices.In order to accurately measure the concerns of investors in the crude oil market,this paper uses the principal component analysis method to weight the search volume to construct the Google Search Volume Index(GSVI).Based on GSVI,a four-variable(supply,demand,investor concern,price)structural vector autoregressive(SVAR)model was constructed to study the influence of investors on WTI crude oil price and interpretation ability.The results show that investor attention has a significant negative impact on WTI crude oil prices,and the contribution to oil price fluctuations reaches 15.18%.This part of the study reveals and quantifies the impact of investor concerns on oil prices.The fifth chapter studies the role of regime switching in the prediction of oil price volatility.In order to determine whether the consideration of regime switching in forecasting models can improve the forecasting accuracy of oil price volatility,this paper evaluates the forecasting performance of three single-mechanism GARCH(including GARCH,GJR-GARCH and EGARCH)models and two regime-switching GARCH(including MMGARCH and MRS-GARCH)models using six loss functions and the model confidence set method.The results show that considering the regime switching does not significantly improve the prediction accuracy of oil price volatility.At the same time,there is no absolute optimal volatility forecasting model.The forecasting performance is affected by the sample period,data frequency and evaluation criteria.This finding challenges the significance of mechanism conversion in the prediction of oil price volatility and provides a reference for the selection of models when forecasting crude oil market volatility.The sixth chapter studies the impact of oil price shocks on stock market in oil importing countries and oil exporting countries based on SVAR model,the Dynamic Panel Regression model and Dynamic Panel Quantile Regression model.In order to deal with oil price fluctuations and endogenous problems in the stock market,predecessors often use SVAR models to conduct research based on time series,and the conclusions obtained from these studies are often not universal.Different from the existing literature research,this paper first decomposes the exogenous oil price shock through the SVAR model and then uses the Dynamic Panel Regression and Dynamic Panel Quantile Regression model to study the impact of oil price shocks on oil stocks in oil importing and exporting countries.The results of this study not only reveal the impact of oil price shocks on oil stocks in oil importing countries and oil exporting countries but also reveal whether the impact varies with the state of the stock market.