Optimization Of China 's Financial Risk Monitoring And Warning Methods And BP Neural Network Model

In the long course of China’s economic construction, national fiscal always plays a pivotal role. China today has entered a critical period of economic transition, this is not only the hard time of economic development, but also the sensitive period of fiscal stability. During this period, our country face more and more contradictions and conflicts on the national interests. As a last defenseline of economic security, fiscal face greater challenges. Fiscal risk itself has the characteristics of vague and abstract, the fiscal system in China’s history has its specificity problem, which made the risk of fiscal crisis largely beyond its appeared. Face with the current situation, it is necessary to build an applicable fiscal risk monitoring and early warning system. The core of the monitoring and early warning system is the monitoring and early warning method. A scientific and applicable monitoring and early warning method can directly measuring our country’s fiscal picture, giving an advanced warning of financial crisis for the future, which will greatly help the government make a correct response before the fiscal crisis.This paper consists of the following parts. First of all, analyses the formation and transmission mechanism of fiscal risks. Analyzing and interpretation the sources and unstable factors of public fiscal risk from the different angle. Second, investigating both at home and abroad literature of fiscal risk monitoring and early warning methods, reviewing the history of the methods. According to the characteristics which made the methods useful in specific field, divide the methods in three parts: finance, macroeconomics and fiscal. Again, according to the collected monitoring and early warning method,on the one hand, based on the characteristics analyzing the method from multi-aspects, on the other hand, exploring the applicability of various methods from the essential theory of the methods. Use the scale evaluation methods, to evaluate the overall monitoring and early warning methods and selectthemost suitable method of our country’s public fiscal risk monitoring and early warning. Through the comparison, it is concluded that BP neural network model is the most suitable model for China’s fiscal risk monitoring and early warning method.Finally, using the BP neural network model to construct public fiscal risk monitoring and early warning system, constructingcompletelyfiscal risk monitoring and early warning circuit loop with indicators, using data from 1995 to 2014 in China to trial and verify the method.To analyze the empirical results, rethink the study,and trying to put forward constructive policy suggestions, forming the sustainability of the follow-up study.