The Empirical Research Of The Price Risk In Chinese Cotton Futures Market

By 2015, China’s cotton futures market has experienced a significant ups and downs: in the first half, Zhengzhou cotton price has up to 13640 Yuan/ton from 13000 Yuan/ton in the early January, rose 4.92 percent. Then the price fell to 11395 Yuan/ton, a new lowest price in recent years. In the face of this situation, the industry engaged in a fierce debate on how to control the price risk in Chinese cotton futures market effectively. However, the academia has rarely uttered his voice, almost no scholar make quantitative research for the price risk in Chinese cotton futures market.Under this background, this paper established different price risk models according to nonparametric method, parametric method and semi-parametric method to make empirical analysis on Chinese cotton futures market. We attempt to measure the price risk and choose the most suitable method. This is not only beneficial to guide the investors to invest rationally and avoid risk effectively, but also has a broad and profound social significance for improving the system and the market mechanism in Chinese cotton futures market.The semi-parametric VaR model is the key point in this paper. We introduced the normal distribution, student t distribution and GED distribution into the GARCH models to measure the VaR, which made it more relevant and achieved good fitting results. Finally tested the empirical results through the Kupiec test to support the results.The empirical results show that:at the 95% confidence level, the historical simulation is better than GARCH-VaR under parametric method which can only meet the examination under the t-distribution and GED distribution. In the semi-parametric method, GARCH-VaR can deal with the tip peak and fat tail characters more effective...