有限样本下煤硫分典型值的自举模拟取样估计

在对无烟煤硫分基础统计学特征分析的基础上，采用内核密度估计对硫分数据多态性进行了研究，根据非正态分布的特点，使用Bootstrap计算机模拟取样方法对试验样本值模拟重复取样，以多次Bootstrap模拟取样的均值与标准误差作为无烟煤硫分有限样本典型值及标准误差的稳健估计，实践证明，Bootstrap模拟取样估计对无烟煤硫分典型值的估计是有效的.

Estimation for typical value of sulfur content in anthracite by Bootstrap resampling method in limited sample case

In limited sample case,typical value of sulfur content in anthracite was determined by kernel density estimation coupled with Bootstrap resampling techniques.The multimodality of data distribution was investigated by kernel density estimation;due to its binomial distribution,Bootstrap resampling method was introduced to acquire the typical value of sulfur content in anthracite,in which large numbers of simulated datasets were generated by computer from existing measurements,so that approximate mean and confidence intervals of the derived quantities may be obtained by direct numerical evaluation.It is clearly demonstrated that resampling techniques are powerful tools in analyzing variability and uncertainty in non-normal data sets.