Permutation tests for equality of distributions in high‐dimensional settings

By Peter Hall, N. Tajvidi. 2002

158 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability, Statistical Learning Theory

How researchers see this paper

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

In Artificial Intelligence

In Statistics and Probability

In Statistical Learning Theory

Citations in the corpus

A Test of Relative Similarity For Model Selection in Generative Models (2015, background): “Other two sample tests are not easily extendable to relative tests (Rosenbaum, 2005; Friedman and Rafsky, 1979; Hall and Tajvidi, 2002).”
FastMMD: Ensemble of Circular Discrepancy for Efficient Two-Sample Test (2014, background): “This task, however, is very difficult and challenging in practice since the underneath distribution information are generally unknown apriori (Bickel, 1969; Friedman & Rafsky, 1979; Hall & Tajvidi, 2002; Biau & Gyofi, 2005).”
Testing Hypotheses by Regularized Maximum Mean Discrepancy (2013, methodology): “MMD [9] was experimentally shown to outperform many traditional two-sample tests such as the generalized Wald-Wolfowitz test, the generalized Kolmogorov-Smirnov (KS) test [2], the Hall-Tajvidi (Hall) test [12], and the Biau-Györf test.”
A Kernel Method for the Two-Sample-Problem (2006, background): “Another recent test was proposed in [15] ( Hall): for each point from p, it requires computing the closest points in the aggregated data, and counting how many of these are fromq (the procedure is repeated for each point from q with respect to points fromp).”
Testing for Homogeneity with Kernel Fisher Discriminant Analysis (2007, methodology): “We performed the same experiments for the Maximum Mean Discrepanc y d the Tajvidi-Hall test statistic (TH, [13]).”
Integrating structured biological data by Kernel Maximum Mean Discrepancy (2006, background): “The test statistic is costly to compute; Hall and Tajvidi (2002) consider only tens of points in their experiments.”
Interval-wise testing for functional data (2017, methodology): “The latter approach relies instead on permutation or bootstrap computational intensive techniques (Hall and Tajvidi 2002; Cardot et al. 2007; Hall and Van Keilegom 2007).”
Two-sample tests in functional data analysis starting from discrete data (2007, methodology): “For these reasons, the conclusions we reach for two-sample hypothesis tests have direct implications for k-sample problems. Recent work on two-sample hypothesis tests in nonparametric and semiparametric settings includes that of Louani (2000), Claeskens, Jing, Peng and Zhou (2003) and Cao and Van Ke”
Nonparametric multivariate rank tests and their unbiasedness (2012, background): “Other authors have constructed various permutation tests: Bickel [3], Brown [4], Hall and Tajvidi [14], Neuhaus and Zhu [26], Oja [28], Wellner [42] and others.”
A Kernel Two-Sample Test (2012, methodology): “Another distribution-free test (Hall) was proposed by Hall and Tajvidi (2002): for each point fromp, it requires computing the closest points in the aggregated data, and counting how many of these are from q (the procedure is repeated for each point fromq with respect to points fromp).”
Least-squares two-sample test (2011, methodology): “MMD was experimentally shown to outperform other homogeneity tests such as the generalized Kolmogorov-Smirnov test (Friedman & Rafsky, 1979), the generalized Wald-Wolfowitz test (Friedman & Rafsky, 1979), the Hall-Tajvidi test (Hall & Tajvidi, 2002), and the Biau-Györfi test (Biau & Györfi, 2005).”