| Since the new century, the price of international crude oil has shown a different law comparing with that of the past. The fluctuation in price has been more frequent and fierce, and the phenomenon can not be reasonably explained by traditional economic theory. Under such circumstance, it plays a great role to analyze the reason of the fluctuation in crude oil price and predict the future price of international crude oil. In This paper, both qualitative and quantitative methods are used to analyze the problem. We use qualitative method to determine the main factors that lead to the fierce fluctuation of international crude oil price. Also, quantitative method is used to predict the price of international crude oil under different models. This text will choose the best one from these models by comparing their advantages and disadvantages.The conclusions of the research are as follows:(1) The relation of supply and demand of international crude oil is the most important and long-effecting factor. While in short terms, rare events, gusty political events, unusual climate, speculation factors on future markets and other factors will have short effect on the oil price by impacting or changing investors' prediction towards the relation of supply and demand. (2) The factors that influence the price of international crude oil become more and more and the reasons of the jumps in price become more complex.(3) ARIMA model has certain feasibility on prediction of short-time international crude oil, but has little effect on the prediction of long-time international crude oil price. (4) The volatility of International crude oil's monthly logarithm income clusters. Using the parameter models, we obtain a conclusion that comparing with that of ARIMA model, the prediction accuracy of AR-GARCH model has some improvement. (5) Comparing with AR-GARCH model and ARIMA model, nonlinear BP nerve net model has an obvious advantage on the prediction of international crude oil price. |
PDF Full Text Request