The Bp Neural Network In The Basic Endowment Insurance Fund Pays The Application Of The Risk Early Warning

In recent years, due to the basic endowment insurance fund system design is not perfect, the proportion of payment is unreasonable and the substitution rate is too high and other problems, our country basic endowment insurance is facing the crisis of insufficient capacity to pay. Coupled with the arrival of an aging population, the proportion of the aged population rising, the basic old-age insurance fund management of the hitherto unknown challenge. The basic old-age insurance fund is foundation and guarantee to the common survival, the risk is bound to affect its ability to pay, which will threaten to people’s survival and development, and even to threaten the stability of the whole society. Consequently, our country regulatory authorities should be on the basic old-age insurance fund risk early warning and prevention as a major research topic. The arrival of an aging population and the corruption case a lot of occurrence problems which remind our country government regulatory departments, through the establishment of risk early warning and prevention mechanism, the basic old-age insurance fund risk prevention measures, this can strengthen the government the basic old-age insurance fund supervising and managing, but also can avoid the basic old-age insurance fund assets are damaged, which can ensure that the fund beneficiaries legitimate interests from infringement. Therefore, the urgent need to establish a scientific, reliable prediction method based on monitor and control mechanism on the basic old-age insurance funds to carry out early-warning, ensure the safety of fund, in order to reduce the risk of the pension payments, to prevent and resolve payment risks occur on the lives of ordinary people adversely affected, ensure national economy stable and healthy development the building of a harmonious society has important practical significance.The author is trying to be commonly used in early warning of financial crisis in the BP neural network is introduced into the basic endowment insurance fund risk early warning management, in order to provide a new early warning tool and method. First of all, in the consolidation on the basis of relevant literature, from the economic theory of early warning and endowment insurance funds on the basis of theory, this research based on the basic endowment insurance fund early warning system; Secondly, from the operation of the system, the system design and the environment three aspects discussed in detail the basic endowment insurance fund in the management of possible risks in the process,and on this basis to establish a set of basic old-age insurance fund risk early warning index system; Thirdly, using Matlab7.81anguage, with the help of his classmates were prepared to realize BP neural network procedures, based on BP network of the basic endowment insurance fund risk early warning model; fourthly, collected between1996to2008in Shanghai city the basic old-age insurance fund historical data for a total of13samples of the model was repeated training and learning, made the error rate is only3.86%the predicted results, verify that the model has good fitting degree; Lastly, based on the international and domestic experience of basic endowment insurance fund risk indicator set five alarm output intervals.Two aspects of the innovation of the thesis:firstly, from the operation of the system, design and environment three respects set up the basic pension insurance of the warning index system and warning indicators; secondly, according to the international and domestic experience by dividing the warning interval can be more truthful responses of basic old-age insurance fund payment capability of live. And this method can be applied in the nationwide, with high feasibility and practical values, the related management can make reference to the police the output to assist decision-making old-age insurance fund risk early warning management.