Research On Method Selection Of PPP Project For Pension Real Estate

At present,the number of aging population is increasing day by day.It is expected that the number of aging population in China will enter an accelerated period in the next 20 to 30 years.The traditional family pension system is overburdened,the supply of socialized pension system is insufficient,and the government’s financial resources alone can’t meet the huge investment of pension system,so other pension modes are urgently needed.At present,under the guidance of relevant policies,China is gradually promoting and applying PPP mode,and has achieved good results.The introduction of PPP financing mode into the pension real estate industry can combine the advantages of the government and the private sector,and effectively solve the problems of capital,technology,management and other aspects in the construction of pension real estate.However,there are some problems in the implementation of the PPP project of pension real estate,such as the choice of mode in the operation of PPP project of pension real estateFirst of all,this paper deeply analyzes the relevant theories of PPP mode and the related concepts,attributes and characteristics of pension real estate,explains the principle of grey correlation theory,selects the model suitable for this paper,i.e.BP neural network,based on the comparison and selection of PPP mode selection methods,and explains its concept and learning algorithm.Secondly,the paper analyzes the current situation of PPP project mode selection of pension real estate.Based on the attributes and characteristics of pension real estate project and previous scholars’ research,it is proposed to select 10 influencing factors closely related to pension real estate project,including economic belt,urban classification,GDP,per capita disposable income of urban residents,elderly population(over 60 years old),total construction area,total investment,cooperation period,industry and return mechanism of the project.The octopus collector is used to collect the PPP projects of pension real estate in the PPP project database of the Ministry of finance.Combined with China statistical information network,the data indicators required for the factors to be selected are found out The grey correlation analysis method is used to analyze the relationship between the 10 factors to be selected and the project operation mode The results show that there is a good correlation between the factors to be selected and the project operation mode,which provides theoretical support for the next operation mode selection.Then,on the basis of the theory in Chapter 2,a three-layer BP neural network prediction model is constructed by selecting activation function,setting network parameters and determining neural network structure.According to the data collected and verified in Chapter 3,10 influencing factors are taken as input indexes,operation mode as output indexes,and data are input into the established network for training and testing after processing,and the test results are concluded The coefficient of fruit determination R2＝0.9173,approaching to 1,the effect of network model is better.Finally,taking the PPP project of Jinan high tech Zone East Hospital and summer care center(phase 1)in Shandong Province as an example,input the index data of various influencing factors into the trained neural network model,and according to the interval value defined in section 4.2.3,it is suggested that the project adopt BOT modeTo sum up,the research in this paper provides support for decision-makers to make a scientific choice of mode,provides reference and guidance for the choice of PPP mode in the future,and has certain reference significance and value for promoting the smooth operation of pension real estate projects.