中国城市地价波动的幂律特性

Power Laws Features of Land Price Fluctuation in Chinese Cities

研究目的:实证分析中国城市地价波动的幂律特性,并比较幂指数的差异性。研究方法:最小二乘法(OLS),最大似然估计法(MLE),柯尔莫哥洛夫—斯米尔统计量。研究结果:中国城市综合地价、商服地价、住宅地价和工业地价的增长率波动都具有幂律特性,且所估计的幂指数都接近2,但工业地价对应的幂指数明显偏小,这些幂律规律与金融市场的幂律规律具有相似性,而幂指数却存在显著差异性。研究结论:城市地价波动的幂律特性是普适性的规律,能为土地市场进一步的发展完善提供有价值的政策启示。

The purpose of this study is to empirically explore the power-law features of land price fluctuation in Chinese cities, and compare the differences between the power-law exponents. The methods used include Ordinary Least Squares linear regression and Maximum Likelihood Estimation combining with Kolmogorov-Smirnov（KS）statistic. The results find that the fluctuation of growth rate of four different types of land price in Chinese cities, including integrated land price, commercial land price, residential land price and industrial land price, all obey power laws. The estimated exponents for land prices are approximately equal to two, but exponent for industrial land price is smaller. These power laws in land market are similar to those in financial market, yet they have significantly different scaling exponents. We conclude that the power-law features of land price fluctuation in Chinese cities are universal laws, and these statistical regularities can provide valuable policy implications for further improvements in land market.