Neural network economic forecast method based on genetic algorithm

Economic forecasting is a kind of conjecture and estimation future economic phenomena. Proceeding from the history reality development, this paper uses scientific methods to reveal development law various phenomena, point out relationship between their direction possible extent. Based on this, aims predict growth Province C proposes prediction tool for improving BP neural network based genetic algorithm. Forecast data GM11 used as training sample neutral web, performed after completed. This concludes through experimental exploration that outcomes are superior web model improved by legacy The error difference result actual value 3.6%, average less than half time series model, which only 1.4%.