On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other

By H. B. Mann, D. Whitney. 1947

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

An Empirical Study of Metrics to Measure Representational Harms in Pre-Trained Language Models (2023, methodology): “Mann-Whitney U-test does not assume any specific distribution such as normal distribution of samples for calculating test statistics and p-values.”
“I’m sorry to hear that”: Finding New Biases in Language Models with a Holistic Descriptor Dataset (2022, background): “test (Mann and Whitney, 1947) to test the hypothesis that, for two templated sentences A and B with different descriptors, there is an equal likelihood of either sentence to have a higher perplexity than the other.”
Comparing Code Explanations Created by Students and Large Language Models (2023, methodology): “As Likert-scale response data is ordinal, we used the non-parametric Mann–Whitney U test [28] to test for differences in Likert-scale question data between student and LLM code explanations.”
Achieving Human Parity on Automatic Chinese to English News Translation (2018, methodology): “Our implementation differs from this as the clustering has been integrated into Appraise and uses the Mann-Whitney rank test [26] at the same p-level p ≤ 0.”
Physics-Informed Neural Networks for Cardiac Activation Mapping (2020, methodology): “We test this hypothesis with the Mann-Whithney test [31] and obtain a signiﬁcant diﬀerence for all cases, both in activation time ( p 10 − 7 ) and in conduction velocity ( p 0.015).”
Learning to Continually Learn (2020, methodology): “26 × 10 − 8 ; all p-values computed using the Mann-Whitney U test [31]).”
Adversarial Removal of Demographic Attributes from Text Data (2018, methodology): “One-tailed Mann-Whitney U test (Mann and Whitney, 1947) showed the effect is highly signiﬁcant with p e − 12 .”
Revisiting Classifier Two-Sample Tests (2016, methodology): “…tests include the t -test (Student, 1908), which tests for the difference in means of two samples; the Wilcoxon-Mann-Whitney test (Wilcoxon, 1945; Mann & Whitney, 1947), which tests for the difference in rank means of two samples; and the Kolmogorov-Smirnov tests (Kolmogorov, 1933; Smirnov,…”
Temperature models for pricing weather derivatives (2012, methodology): “(Mann and Whitney 1947) between the MSRE of each pair of models.”
The Simple Difference Formula: An Approach to Teaching Nonparametric Correlation1: (2014, methodology): “A mathematically equivalent version of the rank-sum test is the Mann-Whitney U test (Mann & Whitney, 1947).”
Merging chrominance and luminance in early, medium, and late fusion using Convolutional Neural Networks (2015, methodology): “The same test is also known as Mann-Whitney U test, as it was designed independently by Mann and Whitney in 1947 [Mann and Whitney, 1947].”
Financial Stress, Downturns, and Recoveries (2009, methodology): “Then, we employ nonparametric tests based on the work by Wilcoxon (1945) and Mann and Whitney (1947) which tests the null hypothesis that the two groups were drawn from populations with the same median.”
Metagenomic biomarker discovery and explanation (2011, methodology): “The LEfSe algorithm is introduced in overview in the Results section, and Figure 6 illustrates in detail the format of the input (a matrix with n rows and m columns) and the three steps performed by the computational tool: the KW rank sum test [49] on classes, the pairwise Wilcoxon test [50,51] betw”
Measuring the Discriminative Power of Rating Systems (2003, methodology): “The interpretation of AUC as a probability relates to the test statistic of the U-test of Mann and Whitney(1947). If we draw a defaulter with score sD from SD and a non-defaulter with score sND from SND and defineuD,ND as”