Randomization Tests for a Multivariate Two-Sample Problem

Chung, Janet; Fraser, D. A. S.

doi:10.1080/01621459.1958.10501472

Cited by 68 publications

(17 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This analysis was repeated 10,000 times, and the proportions of the random replications that resulted in as many significant proteins as seen in the actual data were reported as the significance levels for the number of proteins. 13,14 Significance of class prediction dependent on unsupervised analysis was based on a Fisher's exact test or 2 test, as appropriate. Relatedness in cytokine expression patterns before and after IL-2 administration was displayed by Eisen's hierarchical clustering applied to proteomics as previously described.…”

mentioning

confidence: 99%

Serum Vascular Endothelial Growth Factor and Fibronectin Predict Clinical Response to High-Dose Interleukin-2 Therapy

Sabatino

Kim‐Schulze

Panelli

et al. 2009

JCO

164

104

View full text Add to dashboard Cite

Purpose High-dose interleukin-2 (IL-2) induces durable therapeutic responses in a small subset of patients with metastatic melanoma and renal cell carcinoma, but simple pretreatment predictors of response have not been identified. Patients and Methods To identify predictive biomarkers of clinical response, sera from patients treated with high-dose IL-2 were collected for analysis using a customized, multiplex antibody-targeted protein array platform that surveyed expression of soluble factors associated with tumor immunobiology. Soluble factors associated with clinical responses were analyzed using a multivariate permutation test, and survival outcomes were determined using Kaplan-Meier and log-rank tests. Results A training set from 10 patients identified 68 potentially relevant soluble factors that were then tested in an independent validation set of 49 patients. Class comparison revealed a cluster of 11 biomarkers that were associated with therapeutic outcome. Vascular endothelial growth factor (VEGF) and fibronectin were identified as independent predictors of response. In particular, high levels of these proteins were correlated with lack of clinical response and decreased overall survival. Conclusion Serum VEGF and fibronectin are easily measured pretreatment biomarkers that could serve to exclude patients unlikely to respond to IL-2 therapy.

show abstract

mentioning

confidence: 99%

Serum Vascular Endothelial Growth Factor and Fibronectin Predict Clinical Response to High-Dose Interleukin-2 Therapy

Sabatino

Kim‐Schulze

Panelli

et al. 2009

JCO

164

104

View full text Add to dashboard Cite

show abstract

“…Multivariate two-sample tests have been of continuous interest to the statistics community. Chung and Fraser (1958) proposed several randomization tests. Bickel (1969) constructed a multivariate two-sample test by conditioning on the empirical distribution function of the pooled sample.…”

Section: Introductionmentioning

confidence: 99%

Ensemble Subsampling for Imbalanced Multivariate Two-Sample Tests

Chen

Dou²,

Qiao

2013

Journal of the American Statistical Association

View full text Add to dashboard Cite

Some existing nonparametric two-sample tests for equality of multivariate distributions perform unsatisfactorily when the two sample sizes are unbalanced. In particular, the power of these tests tends to diminish with increasingly unbalanced sample sizes. In this paper, we propose a new testing procedure to solve this problem. The proposed test, based on a nearest neighbor method by Schilling (1986a), employs a novel ensemble subsampling scheme to remedy this issue. More specifically, the test statistic is a weighted average of a collection of statistics, each associated with a randomly selected subsample of the data. We derive the asymptotic distribution of the test statistic under the null hypothesis and show that the new test is consistent against all alternatives when the ratio of the sample sizes either goes to a finite limit or tends to infinity. Via simulated data examples we demonstrate that the new test has increasing power with increasing sample size ratio when the size of the smaller sample is fixed. The test is applied to a real data example in the field of Corporate Finance.

show abstract

“…Some authors have suggested methods of direct combination applied to the T k j statistics themselves [5,6]. They are more appealing computationally, being O(nM N ) instead of our method's O(nM N log(N )), but they do not avoid problems of differing scale or strong correlation.…”

Section: Relation To Other Testing Proceduresmentioning

confidence: 99%

“…Unlike methods of direct combination, which sum up various test statistics [5,6], our method is invariant to the scale of each term. This is critical when different shape parameters have different physical units and the choice of weighting between them can be arbitrary.…”

Section: Introductionmentioning

confidence: 99%

Hypothesis Testing with Nonlinear Shape Models

Terriberry¹,

Joshi²,

Gerig

2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We present a method for two-sample hypothesis testing for statistical shape analysis using nonlinear shape models. Our approach uses a true multivariate permutation test that is invariant to the scale of different model parameters and that explicitly accounts for the dependencies between variables. We apply our method to m-rep models of the lateral ventricles to examine the amount of shape variability in twins with different degrees of genetic similarity.

show abstract

Randomization Tests for a Multivariate Two-Sample Problem

Cited by 68 publications

References 4 publications

Serum Vascular Endothelial Growth Factor and Fibronectin Predict Clinical Response to High-Dose Interleukin-2 Therapy

Serum Vascular Endothelial Growth Factor and Fibronectin Predict Clinical Response to High-Dose Interleukin-2 Therapy

Ensemble Subsampling for Imbalanced Multivariate Two-Sample Tests

Hypothesis Testing with Nonlinear Shape Models

Contact Info

Product

Resources

About