Empirical Likelihood Method And It’s Application To Chinese Oil Price Trend Forecast

Since the1960’s,due to the needs of economic development, more and more international countries become more and more relying on oil and the non-renewable resources,The world has begun making much competition for this scarce resources. Due to various factors the price of crude oil present ups and downs,and trend of rising, So the oil price fluctuations on the world has a profound impact on the development of national economy.With China’s reform and opening up in the1980s the accession to the WTO in2001, the price of international crude oil market volatility is becoming more and more influential to China’s oil prices,which in turn has a significant impact on the development of China’s economy,So estimates of oil price fluctuations is attracting more and more attention of many scholars.In financial engineering, Black-Scholes formula is a commonly used method of pricing derivatives.However,in recent findings of econometrical data, the parameter representing volatility was originally assumed to be constant, which is now believed time-dependent.Conditionalheteroskedastic models have been largely investigated since Engle [8] introduced his renowned ARCH model. Empirical data shows that, in practice, the innovations are not always normally distributed. In fact, the distributions of the innovations of the underlying assets are unknown and sometimes possess heavier tails than the normal distribution. In light of the evidence, instead of MLE, in the case when normality assumption is violated, we adopt the empirical likelihood method to estimate the parameters emerging in the GARCH models.Therefore, how to apply the empirical likelihood method to estimate the parameters of GARCH model is the focus of this thesis.In this thesis, the structure is as follows:The first chapter,summarizes the method of GARCH model and method of empirical likelihood.It will help us to understand the maximum likelihood estimation in GARCH models and its application,understand the theory of empirical likelihood method and grasp its estimation method.The second chapter is the introduction of ARCH model and GARCH model,and introduce how to build a volatility model,summarizes the steps of building a volatility model through GARCH model.The third chapter is the main part of this thesis,We will applying empirical likelihood method to estimate the GARCH model, and give mathematical derivation.In the fourth chapter,we apply empirical likelihood method using R statistical software to fit the GARCH model and estimate its parameters,and model the oil price volatility.At last,we compare the effectiveness of the empirical likelihood estimate and the maximum likelihood estimate of the GARCH model.In the fifth chapter, discussion is made based on the previous chapters and the estimate results were analyzed.Finally we make a conclusion.