Exact p-values for pairwise comparison of Friedman rank sums, with application to comparing classifiers
详细信息 查看全文
- 作者：Rob Eisinga ; Tom Heskes ; Ben Pelzer ; Manfred Te Grotenhuis
- 关键词：Friedman test ; Exact p ; value ; Rank sum difference ; Multiple comparison ; Nonparametric statistics ; Classifier comparison ; Machine learning
- 刊名：BMC Bioinformatics
- 出版年：2017
- 出版时间：December 2017
- 年：2017
- 卷：18
- 期：1
- 全文大小：1293KB
- 刊物主题：Bioinformatics; Microarrays; Computational Biology/Bioinformatics; Computer Appl. in Life Sciences; Algorithms;
- 出版者：BioMed Central
- ISSN：1471-2105
- 卷排序：18
文摘
BackgroundThe Friedman rank sum test is a widely-used nonparametric method in computational biology. In addition to examining the overall null hypothesis of no significant difference among any of the rank sums, it is typically of interest to conduct pairwise comparison tests. Current approaches to such tests rely on large-sample approximations, due to the numerical complexity of computing the exact distribution. These approximate methods lead to inaccurate estimates in the tail of the distribution, which is most relevant for p-value calculation.