Application of GA-SVM time series prediction in tax forecasting

Forecasting the tax gross exactly is significant to carry on the macroscopic regulation efficiently under the market economy. Conventional linear macroscopic economic model is very difficult to hold non-linear phenomena in economic system, thus the tax forecasting error will increase. Support vector machine (SVM) has been successfully employed to solve regression problem of nonlinearity and small sample. However, the application for tax forecasting is neglected. Based on regression arithmetic of SVM, support vector machine with genetic algorithm (GA-SVM) is proposed to forecast tax, in which genetic algorithm (GA ) is used to determine the training parameters of support vector machine. The experimental results indicate that the proposed GA-SVM model can achieve great accuracy under the circumstance of small training data.