Research On Statistical Monitoring And Early Warning Of Poverty Caused By Serious Diseases In Post-poverty Alleviation Era

By 2020,China will have lifted its poverty-stricken population out of poverty,and all poverty-stricken counties will be lifted off their caps to eliminate overall regional poverty,which means that China has entered a post-poverty alleviation era.It is the inevitable requirement of poverty management in the post-poverty alleviation era to change the thinking of poverty reduction and change the governance mechanism based on poverty alleviation to the governance mechanism based on poverty prevention.Risk statistics,monitoring and early warning are important measures to protect people’s better life in the post-poverty alleviation era by actively monitoring and intervening risks that may lead to poverty,blocking them at key points and doing a solid and detailed job of risk prevention and resolution.On the basis of reviewing the research results of scholars at home and abroad,this paper,based on the theory of health,poverty,risk and early warning,conducts the research on the statistical monitoring and early warning of the serious disease caused by poverty in the postpoverty alleviation era according to the mechanism of the major disease caused poverty and the coping strategies of families facing the disease risk.Using the data from the 2018 China Household Dynamics Tracking Survey database of China Social Science Survey Center of Peking University,according to the factors affecting catastrophic health expenditure,combined with the availability of data,three categories of 11 indicators affecting the occurrence of catastrophic health expenditure were screened out.According to whether the family medical expenses exceed 12% of the total household income to determine whether the family happened catastrophic health expenditure,the data collected in accordance with the one thousand one hundred percent cross validation method divided into training set and test set,using the grid search method adjusting the parameters of training support vector machine(SVM)model of the optimal kernel function and kernel parameter,To find the highest classification accuracy in the cross-validation sense to predict whether families are at significant disease risk;Referred to the multi-dimensional poverty indicators developed by the United Nations Development Program,made appropriate modifications according to the research purpose,and added the indicators with the nature of resisting the risk of poverty caused by serious diseases: Assets,labor burden of population,social resources to build a multi-dimensional relatively poor identification system,major disease risk based on the family and the family is relatively poor deprived dimension to build a serious illness,poverty statistics monitoring and early warning mechanism to monitor the risk of a serious illness poverty family,when the monitoring results to achieve a warning level,the information will be passed to the government,this part of the crowd began to concern,When the monitoring results reach the corresponding level of intervention,according to the real situation of the poor group,appropriate means are selected to intervene in this part of the group.Through the research,the main conclusions are as follows :(1)The optimal kernel function and kernel parameters are found by using the ten-fold cross validation algorithm,so that the support vector machine can find the highest classification accuracy in the sense of cross validation.(2)The accuracy of SVM in predicting the occurrence of catastrophic health expenditure is within a reasonable error range,which has theoretical and practical significance.(3)based on the identified families with major disease risk and multidimensional relative poverty index build poverty monitoring early warning mechanism,have a significant risk of disease family according to the results of multidimensional poverty deprived is divided into five grades,and proposed the corresponding countermeasure,to effectively prevent serious poverty and achieve sustainable poverty provides reference.