An Empirical Study On The Futures Program Trading Model

Program trading is a computer program that connected via the terminal server based personal computer and trading automatically. It is quite common in the developed securities markets because its huge scale can ensure the liquidity of huge amounts of money like USA, the laws and regulations is relatively perfect, and its transaction cost is low, these factors ensure program trading has become popular in the developed securities market. But Chinese securities market is without these factors, the foreign program trading is not completely into China. At the same time, program trading of China native also started quietly, especially in Chinese futures market, because speculate on a fall is advantage, and futures price by the spot the impact of supply and demand is very large, the influence factors is less than stock. Investors are more likely to do fundamental analysis and technical analysis. At present, a lot of program trading is mainly based on the average index of price, trends and shocks are the two commonly used methods. The methods is simple and practical, the majority of the industry also believe that the more simple strategy is a better strategy.The purpose of this paper is that attempt to explore a better program trading strategies for the agricultural product futures market. According to the characteristics of Chinese palm oil and other agricultural products futures, through comparative study of several trading strategy, choose a better strategy, hope to have the instruction function to the investors.The innovation point of this article is build model by using the price fluctuations, but the most is build model use of spread arbitrage, and the Kelly formula applied to arbitrage model.The research method of this paper is the empirical analysis and comparative analysis, empirical research extract the data of China Dalian Commodity Exchange futures, established the GARCH model, arbitrage model of palm oil and soybean oil, causal relationship model of palm oil and soybean oil. The results of a comparative analysis is that the arbitrage model of palm oil and soybean oil is the best, followed by causal relationship model of palm oil and soybean oil, GARCH model effect is poor.