中国城市人口规模分布规律研究

Research on the Population Size Distribution across Cities in China

文章利用中国1995、2000、2005、2010和2015年城市人口规模数据,通过多参数模型的比较与检验,对中国城市人口规模分布与变化规律进行实证分析。研究结果显示,中国城市人口规模分布并不遵循幂律分布,不适合采用帕累托分布或齐夫定律分析;与简单的对数正态分布相比,中国城市人口规模分布更符合双帕累托对数正态分布,且拥有对数正态分布的中间主体,上下尾部更接近幂律分布,从2000年开始上尾比下尾更厚重;从模型拟合变化来看,过去20年里的5个年份不同人口规模城市的增长速度不同,发散式增长显著,与双帕累托对数正态分布模型拟合的最佳分布或适度分布相比,2000年后处于上尾的大中城市的数量和人口规模增长迅速,处于下尾的中小城市数量多但人口规模增长动力相对不足。这些研究结论有助于检视当前中国的城市人口规模政策,并对新政策的制定提供决策支持。

Based on city population size data of China in 1995,2000,2005,2010 and 2015,the paper analyses empirically both size distribution and dynamic pattern through model comparison and tests.The result shows that Chinese cities' population size distribution is not appropriate to apply the Zipf's Law and Pareto distribution.Double Pareto lognormal(DPLN) distribution provides a better fit to actual city sizes than the simple lognormal(LN) distribution.The DPLN has a lognormal body and features a power law in both the lower and the upper tail.Since 2000,the upper tail of China's city size distribution has become much thicker than the lower tails.According to the dynamics of empirical models,in the past 20 years,the growth rate of different scale cities is different,and the pattern of divergent growth is notable.However,the scale development of big cities is restrained.The size and number of big cities in China is relatively small,comparing with the best distribution or moderate distribution.These conclusions will contribute to review current urban development policies,and provide support to new policies.