Nonparametric procedures for partially paired data in two groups

Solomon W. Harrar; Merga B. Feyasa; Eshetu Wencheko

doi:10.1016/j.csda.2019.106903

Nonparametric procedures for partially paired data in two groups

Solomon W. Harrar, Merga B. Feyasa, Eshetu Wencheko

Dr. Bing Zhang Department of Statistics

Research output: Contribution to journal › Article › peer-review

5 Scopus citations

Abstract

A fully nonparametric method is developed for comparing samples with partially paired data. Partially-paired (correlated) data naturally arise, for example, as a result of missing values, in incomplete block designs or meta analysis. In the nonparametric setup, treatment effects are characterized in terms of functionals of distribution functions and the only assumption needed is that the marginal distributions are non-degenerate. The setup accommodates binary, ordered categorical, discrete and continuous data in a seamless fashion. The use of nonparametric effects addresses the Behrens–Fisher problem from the nonparametric point of view and allows construction of confidence intervals. Although the nonparametric methods are mainly asymptotic, methods for small sample approximations are also proposed. Size and power simulation results show numerical evidence of favorable performance of the nonparametric methods. The new nonparametric method has overwhelming power advantage when treatment effects are in the shape of the distributions and they perform comparably well with parametric methods for location-type alternatives. Data from randomized trials in public health are used to illustrate the application of the method.

Original language	English
Article number	106903
Journal	Computational Statistics and Data Analysis
Volume	144
DOIs	https://doi.org/10.1016/j.csda.2019.106903
State	Published - Apr 2020

Bibliographical note

Funding Information:
The authors are grateful to the associate editor and the three anonymous referees for critically reading the original version of the manuscript and making valuable suggestions that led to substantial improvements. The authors are also thankful to the editor for the orderly handling of the manuscript. The authors would like to express their gratitude to Prof. Derek S. Young for his willingness and time to read through the final version of manuscript and make editorial suggestions. Part of the the research of Merga B. Feyasa was done while holding a visiting student positions at the University of Kentucky (UK). He is thankful to the Department of Statistics at UK and the College of Natural and Computational Sciences of Addis Ababa University for providing support for the visits.

Publisher Copyright:
© 2019 Elsevier B.V.

Keywords

Interaction
Missing data
Nonparametrics
Paired comparison
Rank methods
Relative treatment effect

ASJC Scopus subject areas

Statistics and Probability
Computational Mathematics
Computational Theory and Mathematics
Applied Mathematics

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.csda.2019.106903

Cite this

@article{466d01334c454e949e77fd03e7302198,

title = "Nonparametric procedures for partially paired data in two groups",

abstract = "A fully nonparametric method is developed for comparing samples with partially paired data. Partially-paired (correlated) data naturally arise, for example, as a result of missing values, in incomplete block designs or meta analysis. In the nonparametric setup, treatment effects are characterized in terms of functionals of distribution functions and the only assumption needed is that the marginal distributions are non-degenerate. The setup accommodates binary, ordered categorical, discrete and continuous data in a seamless fashion. The use of nonparametric effects addresses the Behrens–Fisher problem from the nonparametric point of view and allows construction of confidence intervals. Although the nonparametric methods are mainly asymptotic, methods for small sample approximations are also proposed. Size and power simulation results show numerical evidence of favorable performance of the nonparametric methods. The new nonparametric method has overwhelming power advantage when treatment effects are in the shape of the distributions and they perform comparably well with parametric methods for location-type alternatives. Data from randomized trials in public health are used to illustrate the application of the method.",

keywords = "Interaction, Missing data, Nonparametrics, Paired comparison, Rank methods, Relative treatment effect",

author = "Harrar, {Solomon W.} and Feyasa, {Merga B.} and Eshetu Wencheko",

note = "Funding Information: The authors are grateful to the associate editor and the three anonymous referees for critically reading the original version of the manuscript and making valuable suggestions that led to substantial improvements. The authors are also thankful to the editor for the orderly handling of the manuscript. The authors would like to express their gratitude to Prof. Derek S. Young for his willingness and time to read through the final version of manuscript and make editorial suggestions. Part of the the research of Merga B. Feyasa was done while holding a visiting student positions at the University of Kentucky (UK). He is thankful to the Department of Statistics at UK and the College of Natural and Computational Sciences of Addis Ababa University for providing support for the visits. Publisher Copyright: {\textcopyright} 2019 Elsevier B.V.",

year = "2020",

month = apr,

doi = "10.1016/j.csda.2019.106903",

language = "English",

volume = "144",

}

TY - JOUR

T1 - Nonparametric procedures for partially paired data in two groups

AU - Harrar, Solomon W.

AU - Feyasa, Merga B.

AU - Wencheko, Eshetu

N1 - Funding Information: The authors are grateful to the associate editor and the three anonymous referees for critically reading the original version of the manuscript and making valuable suggestions that led to substantial improvements. The authors are also thankful to the editor for the orderly handling of the manuscript. The authors would like to express their gratitude to Prof. Derek S. Young for his willingness and time to read through the final version of manuscript and make editorial suggestions. Part of the the research of Merga B. Feyasa was done while holding a visiting student positions at the University of Kentucky (UK). He is thankful to the Department of Statistics at UK and the College of Natural and Computational Sciences of Addis Ababa University for providing support for the visits. Publisher Copyright: © 2019 Elsevier B.V.

PY - 2020/4

Y1 - 2020/4

N2 - A fully nonparametric method is developed for comparing samples with partially paired data. Partially-paired (correlated) data naturally arise, for example, as a result of missing values, in incomplete block designs or meta analysis. In the nonparametric setup, treatment effects are characterized in terms of functionals of distribution functions and the only assumption needed is that the marginal distributions are non-degenerate. The setup accommodates binary, ordered categorical, discrete and continuous data in a seamless fashion. The use of nonparametric effects addresses the Behrens–Fisher problem from the nonparametric point of view and allows construction of confidence intervals. Although the nonparametric methods are mainly asymptotic, methods for small sample approximations are also proposed. Size and power simulation results show numerical evidence of favorable performance of the nonparametric methods. The new nonparametric method has overwhelming power advantage when treatment effects are in the shape of the distributions and they perform comparably well with parametric methods for location-type alternatives. Data from randomized trials in public health are used to illustrate the application of the method.

AB - A fully nonparametric method is developed for comparing samples with partially paired data. Partially-paired (correlated) data naturally arise, for example, as a result of missing values, in incomplete block designs or meta analysis. In the nonparametric setup, treatment effects are characterized in terms of functionals of distribution functions and the only assumption needed is that the marginal distributions are non-degenerate. The setup accommodates binary, ordered categorical, discrete and continuous data in a seamless fashion. The use of nonparametric effects addresses the Behrens–Fisher problem from the nonparametric point of view and allows construction of confidence intervals. Although the nonparametric methods are mainly asymptotic, methods for small sample approximations are also proposed. Size and power simulation results show numerical evidence of favorable performance of the nonparametric methods. The new nonparametric method has overwhelming power advantage when treatment effects are in the shape of the distributions and they perform comparably well with parametric methods for location-type alternatives. Data from randomized trials in public health are used to illustrate the application of the method.

KW - Interaction

KW - Missing data

KW - Nonparametrics

KW - Paired comparison

KW - Rank methods

KW - Relative treatment effect

UR - http://www.scopus.com/inward/record.url?scp=85076922555&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076922555&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2019.106903

DO - 10.1016/j.csda.2019.106903

M3 - Article

AN - SCOPUS:85076922555

SN - 0167-9473

VL - 144

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

M1 - 106903

ER -

Nonparametric procedures for partially paired data in two groups

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this