重特大事故的离散幂律分布——以煤矿事故为例

Discrete power-law distribution of major accidents—a case study of coalmine fatalities in China

研究近年来我国每起煤矿重大和特别重大事故死亡人数统计数据,建立其概率分布模型。基于极大似然法和柯尔莫哥洛夫-斯米尔诺夫检验方法(K-S方法)估计我国煤矿重大和特别重大事故离散幂律分布的参数,并采用Monte Carlo方法随机生成大量的检验样本,对其进行K-S方法拟合优度检验。结果表明在统计学意义上我国煤矿重大和特别重大事故死亡人数比高斯分布的值大得多,服从标度指数为2.72的离散幂律分布。

The serious fatal coalmine accidents in China in recent years were analyzed to construct a reasonable probabilisfic distribution model. The scaling parameter for discrete power-law distribution in the serious fatalities data was estimated, based on a maximum likelihood method and the Kolmogorov-Smirnov statistic （K-S method）. And a large number of random data drawn from the discrete power-law distribution that best fit those data with the Monte Carlo method were used to make a goodness-of-fit test that compared the fatalities data to the hypothesized distribu- tion, based on the K-S method. The results showed that the serious coalmine fatalities in China were greater than a truncated normal distribution with the same sample mean and standard error, following the discrete power-law distri- bution with scaling exponent 2. 72, statistically.