Nonparametric Autoregressive Model Of Chinese Population Rate

Population is the bottleneck of restricting our country society and the economic development which had already caused close attention of Communist Party of China and Chinese government, is also the important thesis that population experts, sociologist and economists have been researching on. Building appropriate population model is the foundation of population forecast, controlling and supervision, at the same time, is one of the hotspot problem of current population research. The traditional population growth forecast model cannot ideally catch the nonlinear characteristic of our country's population growth rate, but nonparametric methods not only overcome linearity model's deficiency but also modeling flexibly, so they become the most important methods of studying nonparametric model. The text establishes the nonparametric autoregressive forecast model on our country's population growth rate.Firstly, This paper simply establishes the linear autoregressive 2-D model on our country population growth; secondly, we establish nonparametric autoregressive model which utilize kernel estimation and local linear estimation respectively, namely the nonparametric autoregressive 1-D model; the results of fitting and forecast indicate that local linear estimation is more advantageous than kernel estimation .Finally, after the comparison of t the linear autoregressive 2-D model and he nonparametric autoregressive 1-D model about our country population growth, the computed results show that the nonparametric autoregressive model can preferably reflect the nonlinearity characteristic of our country population growth rate data than linear autoregressive model .