函数方差检验方法的应用  

Application of Function Variance Testing Method

作　　者：徐延新[1,2] 刘黎明[1] Xu Yanxin Liu Liming(Statistic School, Capital University of Economics and Business, Beijing 100070, China Department of Mathematics, Harbin Norm University, Harbin 150025, China)

机构地区：[1]首都经济贸易大学统计学院,北京100070 [2]哈尔滨师范大学数学科学学院,哈尔滨150025

出　　处：《统计与决策》2017年第9期23-26,共4页Statistics & Decision

基　　金：全国统计科学研究计划项目(2013LZ13);黑龙江省青年基金资助项目(JJ2016QN0211);北京特大城市研究中心资助项目(TDJD201502)

摘　　要：空气污染问题引起了社会广泛关注,影响空气质量评价的因素有很多,研究这些因素对指标的影响是否都显著尤为重要。文章将函数型方差检验的方法(ANOVA)应用到空气质量的数据中,首先采集到北京市区内分布在不同地方的35个监测站的每天每小时的实时更新数据,对北京市划分的五个行政区域的空气污染成分进行函数方差分析的k样本均值假设检验,进而分析北京市各行政区域空气污染的各成分之间的差异。Air pollution problem has aroused widespread concern, and many factors are involved in air quality assessment. Therefore, it is especially important to study whether these factors affect the air index. This paper applies function-type analysis of variance （ANOVA） testing method to the air quality data. Firstly, the paper collects the real-time updates data per hour per day from 35 monitoring stations located in different areas of Beijing, and conducts hypothesis testing of the sample mean of functional analysis of variance of air pollution component in the 5 administrative regions of Beijing. And then the paper further analyzes the differences of the air pollution component in every administrative region of Beijing.

关 键 词：方差分析 函数型数据 空气污染 假设检验 

分 类 号：O212.4[理学—概率论与数理统计]