Asymmetric Effect Of International Oil Price Shock On Stock Market Based On Bayesian Quantile Model

Quantile regression has gradually become one of the frontiers of Modern Econometrics,solving the optimization problem to describe the distribution characteristics of variable. Compared with the traditional least square estimation method, quantile regression can not only describe the dependent variable distribution trend, but also describe the tail variables. It is a beneficial supplement to the classical mean regression method. In addition, the panel data mixes two-way characteristics of time-series and cross-section data, which includes heterogeneity and common features of the variables in the complex economic environment. That quantile regression model is introduced into the analysis of panel data econometric model to research is a hot research topic in recent years. However, the commonly used methods of parameter estimation(minimizing the effect of individual penalty) may encounter the problem that punished parameter can not be determined because of uncertainty in the process of estimating parameter. Therefore. This paper put forward panel quantile regression model, and Bayesian estimation method was used to solve the parameter estimation in high dimensional integral value for the asymmetric characteristics of financial panel data.According to the distribution of many economic and financial variables of possiblity,there is existence of asymmetry characteristics,thick-tail spikes and so on. First of all, the panel quantile regression model analyzes the structure of the panel data regression, then, the likelihood function is asymmetric distribution of Laplace model and we give the posterior distributions of the model parameters and the Gibbs sampling algorithm. For the asymmetric relation feature variables, this paper construct the comparison method and the Bayesian estimation of random effects panel data model. At the same time, the construction of individual differences in quantile regression model identified the asymmetric relationship between variables. Finally, choosing the industry index data of our country as the research object, we analyzed the effect of international oil prices impact on the return of stock market in China, and the effect was compared between the industries.The results of the study show that: the stock market return rate of China industry has non-central distribution and heavy-tailed characteristics; at the same time, in the context of the financial crisis,the impact of the international oil price fluctuations on stock on China's stock market returns has the features of asymmetry. Specifically, from the stock market point of view, when the stock market return is in good condition, relationship between the stock market of our country and the international oil price fluctuations is positive; in poor condition, there was a negative correlation between the stock market of our country and the international oil price fluctuations in international oil prices is positive; the effect of the negative volatility of oil market on Chinese stock market is weaker than positive effect. From the relevant industries of the stock market, the impact of the international oil price fluctuations on the industry is different to some extent.