EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Residuals and diagnostics for multinomial regression models

Eric A. E. Gerber, Bruce A. Craig

Statistical Analysis and Data Mining: The ASA Data Science Journal · 2024-02 · 17 (1)

DOI: 10.1002/sam.11645 ↗

At a glance

Journal article in Statistical Analysis and Data Mining: The ASA Data Science Journal, published online on 2023-09-29. Written by Eric A. E. Gerber and 1 co-authors. Cited 6 times so far (Crossref); 28 references. The full text is freely available (open access, CC BY 4.0). Funded by: Purdue Research Foundation.

AbstractBibliografiske oplysningerCiteRelated studies

Abstract (original language)

In this paper, we extend the concept of a randomized quantile residual to multinomial regression models. Customary diagnostics for these models are limited because they involve difficult‐to‐interpret residuals and often focus on the fit of one category versus the rest. Our residuals account for associations between categories by using the squared Mahalanobis distances of the observed log‐odds relative to their fitted sampling distributions. Aside from sampling variation, these residuals are exactly normal when the data come from the fitted model. This motivates our use of the residuals to detect model misspecification and overdispersion, in addition to an overall goodness‐of‐fit Kolmogorov–Smirnov test. We illustrate the use of the residuals and diagnostics in both simulation and real data studies.

Funding

  • Purdue Research Foundation

Cite

APA

Gerber, E. A. E., & Craig, B. A. (2024). Residuals and diagnostics for multinomial regression models. Statistical Analysis and Data Mining: The ASA Data Science Journal, 17(1). https://doi.org/10.1002/sam.11645

Vancouver

Gerber EAE, Craig BA. Residuals and diagnostics for multinomial regression models. Statistical Analysis and Data Mining: The ASA Data Science Journal. 2024;17(1). doi:10.1002/sam.11645

BibTeX
@article{Gerber2024,
  author = {Gerber, Eric A. E. and Craig, Bruce A.},
  title = {Residuals and diagnostics for multinomial regression models},
  journal = {Statistical Analysis and Data Mining: The ASA Data Science Journal},
  year = {2024},
  volume = {17},
  number = {1},
  doi = {10.1002/sam.11645}
}
Start for free →