Research On Household Financial Behavior And Its Crisis Prevention Inchina Under The Perspective Of Complexity

Since family finance was proposed in 2006,it has attracted widespread concern in academic circle.Reasonable household financial behavior can smooth the income and consumption of family,promote the growth of household property income,realize the preservation and appreciation of household wealth;inappropriate household financial behavior takes the family into financial crisis,and may endanger macroeconomic development.Therefore,it is of great theoretical and practical significance to make a thorough study on the current situation,development trends,type characteristics and influencing factors of household financial behavior in China,which can not only provide correct guidance for household financial behavior,but also provide important guidance for formulating macroeconomic policies and regulating financial market.Household financial behavior is defined as household asset management behavior in this paper,the existing research results are untangled from the perspective of its connotation,and the main problems existing in the existing research are also reviewed in this paper.In view of the shortcomings of the existing research,household financial behavior and its influencing factors are studied from the perspective of complexity science in this paper.First of all,the current situation and development characteristics of Chinese households and their financial behavior are analyzed.Secondly,the complexity of Chinese household financial behavior,such as openness,dynamism,non-periodicity,self-organization criticality are discussed.Thirdly,a thorough discussion and empirical analysis on the impact path,influence mechanism,result and degree of family financial behavior from macroeconomic and cultural factors and micro-family factors is made.Finally,be aimed at the low financial participation rate and unreasonable asset structure of Chinese households,some policy suggestions are given from the family level,hoping to play a positive role in the effective allocation of financial assets,rational planning of household debt scale,standardized lending channels,and increasing household property income.The latest household survey data of the year 2016 published by CFPS large-scale micro-database are selected,and a variety of empirical research methods are used.Firstly,on the basis of identifying the main types of households in China by fuzzy clustering method,a random forest model is established to classify the types of household financial behavior.Secondly,based on the self-organized criticality of household financial behavior,an extended sand pile model is constructed to evaluate the dynamic stability of the change of household net asset structure.Thirdly,through the establishment of the radial basis function neural network model,an empirical study on the influence approaches,impact and importance degree of various factors on the family financial behavior in China is made.The main conclusions of the empirical study can be summed up in three points.Firstly,according to their characteristics,household financial behavior can be divided into five main types: traditional type,normative type,active type,speculative type and balanced type,and households take real estate investment as the first choice of asset allocation.Secondly,the unreasonable dynamic changes of the household net assets structure will lead to the collapse of the existing stable state of the household,and the household will loss of the household net assets through ceding assets or foreign debt,which can achieve a new stable state.Thirdly,the degree of participation of household financial behavior largely determined by the income and wealth status of Chinese households;the financial literacy of household economic decision makers determines the rationality and yield rate of household financial behavior;the development of regional economy and financial market affects the standardization of household financial behavior.The research of this paper is based on the perspective of complexity science,which is innovative in both theoretical research and empirical research.In the aspect of theoretical research,family financial behavior is defined as a complex behavior system,and its main characteristics and reasons as an open complex system are discussed.In the aspect of empirical study,the fuzzy clustering analysis,stochastic forest model,neural network model and other depth learning and machine learning methods are used to classify and predict the complex system.Compared with the one-way generalized linear model,these methods have higher accuracy.