A taxonomy of multiple regression methods for immunologists

Tyson H Holmes

doi:10.1016/j.jim.2023.113506

A taxonomy of multiple regression methods for immunologists

J Immunol Methods. 2023 Aug:519:113506. doi: 10.1016/j.jim.2023.113506. Epub 2023 Jun 8.

Author

Tyson H Holmes¹

Affiliation

¹ The Human Immune Monitoring Center, Institute for Immunity, Transplantation and Infection, Stanford University School of Medicine, 1651 Page Mill Road, Palo Alto, CA 94304, United States of America. Electronic address: tholmes@stanford.edu.

PMID: 37295711
DOI: 10.1016/j.jim.2023.113506

Abstract

Multiple regression is a powerful tool in the immunologist's toolbox. This paper defines multiple regression, discusses availability and accessibility, provides some additional helpful definitions, treats the topics of transformation and extreme value screening, and establishes the paper's scope and philosophy. Then eleven methods of multiple regression are detailed, giving strengths and limitations. Throughout an emphasis is placed on application to immunological assays. A flowchart to guide selection of multiple regression methods is provided.

Keywords: Cross-sectional data; Distribution-free statistical methods; Immunological assay; Longitudinal data; Multiple regression; Variance heterogeneity.