Price Forecast for Gold Futures Based on GA-BP Neural Network

China´s gold futures have listed in Shanghai Finance Futures Exchange in January 2008. The currency and financial attributes makes the gold futures more important than the general commodity futures. So the gold futures´ research has important theoretical and practical significance. By using genetic algorithms (GA) to optimize the value of linking weight of the BP neural network ,and then constructing the genetic algorithm BP neural network(GA-BP) forecasting model, selecting the New York Futures Exchange´s the April 2008 delivery gold futures contract as the empirical research´s object, using the Matlab software to simulate the forecasting model. The simulation results show that the forecasting model gets properly acceptable results.