Economic analysis on tax model based on BP neural network

It is difficult to accurately analyze forecasting of tax income. This thesis establishes a tax forecasting model based on BP neural network to analyze impacts imposed on tax income by changes of the following economic factors: industrial added value, total investment in fixed assets, total volume of import and export, total volume of fiscal expenditure, resident consumption level, etc. The thesis proves that it is practical to analyze the highly nonlinear tax income forecasting through the neural network. It is of promotion value.