| As the aging of China’s population continues to be serious,some social problems have gradually emerged,such as the increasing burden of old-age security,growing demand for social services for the elderly,and rapidly expanding medical and health consumption expenditures.The prerequisite for solving these social problems is to have a detailed understanding of the health status of the elderly in China and its factors,to further probe into more diverse factors affecting the health of the elderly,and to provide a scientific and reasonable basis for formulating corresponding policies and decisions.This article is based on four longitudinal data(2008,2011,2014,2018)of the Chinese elderly health factors,to discusses the factors of the elderly heath with social-psychology-biological medical model for the analysis as the overall framework,self-rated health status and daily self-care life as independent variables respectively,and five aspects(natural attribute,social economic structure,living habits,the prevalence and related factors of psychological status)as independent variables.Firstly,descriptive analysis was applied to explore the distribution of the health status of the elderly in China.Secondly,the health self-rated status and daily self-care ability of the elderly were taken as dependent variables respectively.The independent variables were selected on the basis of LASSO,and then sequential Logistic regression model and binary regression model were constructed to analyze the factors of the health status of the elderly in China.Furthermore,the factors of health self-rated and self-care ability in daily life were compared and analyzed.Finally,a random forest model was constructed according to the method of machine learning to distinguish and predict the health status of the elderly in China.The main conclusions of this paper are as follow:(1)the older the elderly,the better the self-rated health status,but it is less likely that they have the ability to take care of themselves in daily life.(2)Elderly people with spouses have worse self-rated health status,but are more likely to be able to take care of themselves in daily life.(3)The elderly with a good lifestyle and a positive and optimistic mood have a good self-rated health status and are more likely to be able to take care of themselves in daily life.(4)The elderly who have suffered from serious diseases are less likely to rate their worse health status and have the ability to take care of themselves in daily life.(5)The random forest based on the elderly’s self-care ability in daily life tum out to be good prediction effects.From the perspective of prediction effects,the accuracy rate,precision rate and recall rate of the random forest model are 85.54%,87.56% and 98.78% respectively.(6)There are significant spatial differences in the health level of the elderly population.Based on the above conclusions,this paper gives the following suggestions:(1)Strengthening health education for the elderly.(2)Encouraging the construction of partnership and mutual assistance pastoral bases.(3)Promoting healthy lifestyles.(4)Improving medical and health care services.(5)High-risk groups should be given safety warnings in advance according to the predicted results.(6)Allocate resources to optimize the health status of the elderly in accordance with local conditions. |