Research on Manufacturing Tax Policy Based on Neural Network

Abstract

In recent years, China has formulated a strategic plan to revitalize the development of manufacturing industry, and the state has issued many tax policies to support the development of manufacturing industry. From the perspective of taxes, this paper combs the relevant literature of manufacturing tax policy, and constructs the PMC-AE index evaluation model by adding self coding neural network technology on the basis of traditional PMC, in which 9 primary variables and 34 secondary variables are set to quantitatively evaluate the tax policy of manufacturing transformation and upgrading in Northeast China. It is found that China's current manufacturing tax policy is more reasonable, but there are still deficiencies. We should improve the manufacturing tax policy from the aspects of receptor scope, guarantee incentive form and duration, so as to provide theoretical support for the revision and optimization of the new round of policy.