几种空气质量预报方法的一致性衡量方法研究  

Measure Inter-agreement of Different Air Quality Forecasting Methods

作　　者：侯宜广 赵瑾 

机构地区：[1]徐州市气象局,江苏徐州221003

出　　处：《沙漠与绿洲气象》2009年第4期13-16,共4页Desert and Oasis Meteorology

摘　　要：空气质量预报是复杂的系统工程,也是环境科学研究的热点和难点所在。通过文献综述分析了现有研究的不足,指出现有的研究没有考虑由于偶然性和随机性导致的一致性。基于权重Kappa统计值的方法,在剔除了由于偶然性和随机性造成的一致性的基础上,对3种常用的空气质量预报方法的预测结果的一致性进行了衡量,有利于提高对不同模型预测结果的差异性的认识,对进一步提高空气质量预报的准确率有一定的意义。The forecasting of air quality is a systemic project as well as the focus and nodus of current research. Based on a detailed literature review, this paper firstly analyzed the gap of extant researches, and pointed out the fact that they did not consider the inter-agreement caused by chances and random. Furthermore, using Kappa value and taking out the impact of chance and random, this paper measured the inter-agreement of predictive results of three different air quality forecasting methods. This research would promote the otherness understanding of different forecasting models and to be important to the improvement of forecasting accuracy.

关 键 词：空气质量预报 权重Kappa统计值 内在一致性 表面一致性 偶然一致性 

分 类 号：X16[环境科学与工程—环境科学]