Model Prediction And Analysis Of Population Quantity,Mobility And Structure In Anhui Province

Population issues are related to the rise and fall of the country and the well-being of the people.The analysis of demographic data provides a reliable basis for the government to grasp population trends and formulate relevant policies.In recent years,the national economy and various undertakings of Anhui Province are in the process of rapid development.The province’s GDP has ranked 11 th in China in recent three years,but the population growth situation is becoming more and more serious.According to the data of the seventh census,the total population of Anhui province increased from 59.5004 million in 2010 to 61.0272 million in 2020,with an average annual growth rate of only 0.25%,in a state of low-speed growth;The elderly’s coefficient increased from 10.08% in 2010 to 15.01% in 2020,and the population aging has further deepened;The children’s coefficient increased slowly from17.77% in 2010 to 19.24% in 2020.With the development of society and the influence of factors such as migrant workers,higher education and employment,population mobility is very frequent and tends to be uncertain.Therefore,it is necessary to conduct reasonable analysis and prediction on the total,flow trend and structure of the population.Based on the relevant theories of grey system,BP neural network and time series,this paper establishes a mathematical model to analyze and predict the total,the flow condition and the structure of the population in Anhui Province.In the prediction of total population,GM（1,1）model,two parameter exponential smoothing model and BP neural network model are used to predict the total population respectively.At the same time,in order to improve the accuracy of the prediction,the above three models are modified.When GM（1,1）model is used for prediction,the weakening buffer operator is used to process the original data sequence,and then GM（1,1）model is used for prediction;When the two parameter exponential smoothing method is used for prediction,the variable smoothing coefficient is obtained by combining the equal dimensional innovation algorithm to modify the two parameter exponential smoothing method;In the prediction by BP neural network,the training samples are obtained by BP neural network,and then the error of the training samples is corrected,which is combined with Markov chain to make the prediction results more accurate.Finally,the least square error method is used to predict the combination of the three improved models.The prediction results show that the population is growing slowly,the growth rate is declining from 2021 to 2030,with a trend of negative growth.The prediction of population mobility provides vital preference for the government when it makes relevant policies of talents.Firstly,this paper makes a descriptive statistical analysis on the structure of the population flowing outside Anhui Province,and then uses logistic block model combined with ARMA model to predict the number of floating population.The forecast results show that the growth rate will slow down,reaching 12.45 million by 2025.The prediction of population structure includes children’s coefficient,old age coefficient and so on.Based on Song Jian’s population development equation,this paper forecasts the children’s coefficient and the elderly’s coefficient respectively.In order to improve the accuracy of the prediction,Song Jian’s population development equation is improved from three aspects: first,the population is predicted separately according to gender;Secondly,change the migration population function into the floating population change function;Finally,the mortality function by age and sex is constructed to calculate the mortality.Then the improved model is used to predict the children’s coefficient and the elderly’s coefficient.The prediction results show that while keeping the total fertility rate unchanged,the children’s coefficient will drop to 16.02% and the elderly’s coefficient will rise to 18.22% in 2030.