A Population Prediction Model Based on Variable Weight Coefficients

To deal with the lack of accuracy and consistency in some single models, it took advantage of different algorithms to optimize the BP neural network respectively. Regarding predicting outcomes of basic BP model, GA -BP model and PSO-BP model as the input cells of new BP neural network's learning data, the combination predicting outcome can be got finally. The novel of this combination model lies in the collection of different predicting outcomes of single models together, and the input of neural network together with a combination forecast of the total population. Entering corresponding years simply, combination forecast outcomes can be got at once. It would have no impact because of inaccurate data and unknown corresponding factors. And weight coefficients of combination model change over time is very scientific. Taking advantage of this combination model to predict the total population of China from 2014 to 2020, the results show that total population of China in 2020 is within 1.4 billion. The trend of China's population growth has been effectively controlled by the government and it is of great possibility for the government to control the total population within 1.45 billion in 2020.