基于国内出厂价的航油价格预测与波动风险测度研究

Research on Jet Fuel Price Forecast and Volatility Risk Measurement Based on Domestic Ex-factory Price

我国航油市场起步晚且发展不成熟,缺乏完善的市场规则和监管机制,使得我国航油价格波动性较高,航空公司难以对价格波动带来的风险进行有效管理。因此,从风险测度量化角度出发,构建合适的模型来研究我国航油出厂价的波动风险,为国家政策制定和风险监控提供参考。为了全面分析我国航油出厂价的波动规律、有效识别价格波动风险、预测未来价格,首先建立ARIMA模型,其次借助GARCH族模型探究正态分布、t分布、广义误差分布下的航油出厂价方差波动,检验杠杆效应。研究结论显示我国航油出厂价波动较大,具有“尖峰厚尾”特征;t分布下的GARCH (1,1)模型拟合出厂价对数收益率较好;用TARCH模型拟合发现杠杆效用不明显。基于以上结果,文章建议建立精确有效的风险度量和预警系统、制定长久稳定的航油交易政策以及建立信任和互惠互利的航司与航油供应商合作关系。

Due to the late start and immature development of China's jet fuel market and the lack of perfect market rules and supervision mechanisms,the price volatility of jet fuel in China is relatively high,and it is difficult for airlines to effectively manage the risks brought by price fluctuations.Therefore,from the perspective of risk measurement,a suitable model is constructed to study the volatility risk of ex-factory price of jet fuel in China,so as to provide reference for national policy making and risk monitoring.In order to comprehensively analyze the fluctuation rule of ex-factory price of jet fuel in China,effectively identify the risk of price fluctuation and predict the future price,the ARIMA model was first established,and then the GARCH family model was used to explore the variance fluctuation of ex-factory price of jet fuel under normal distribution,t-distribution and generalized error distribution,and to test the leverage effect.The results show that the ex-factory price of jet fuel in China fluctuates greatly and has the characteristics of"peak and thick tail".The GARCH(1,1)model with t-distribution fits the log return of factory price well.It is found that the leverage effect is not obvious by TGARCH model fitting.Based on the above results,the paper suggests the establishment of accurate and effective risk measurement and early warning systems,the formulation of long-term and stable jet fuel trading policies,and the establishment of trust and mutually beneficial cooperation between airlines and jet fuel suppliers.