Abstract:

Mathematical models are widely used to simulate the development trend of population.In this paper,four mathematical models,named as Malthusian model,Logistic model,linear regression model and Keyfitz model were studied on their capability of modeling the historical development trend of seven population variables of Kunshan,Jiangsu Province from 1985 to 2006.Results showed that,generally,Logistic model was better than other models according to determination coefficients.Then each index was analyzed separately with an aim to explain the differences between simulated and recorded numbers.From the results,it was concluded that:1) logistic model fitted well primarily because it had three parameters while other models had two,and the additional parameter 〖WTBX〗Pm〖WTBZ〗,which should stand for the upper limit of population increasing,did not have real sense in most cases;2) the prediction of future population by the use of mathematical modeling showed unfeasible for the existence of turning points in the development trend of population variables;3) combined with historical information of the research area,turning points could give a good description on the effects of political and economic factors occurred simultaneously;4) the development course of Kunshan should be divided into three stages,the developing stage,the adjustment stage and the overall developing stage,rather than five stages generally accepted previously.

Key words: population variable/mathematical model/turning point