Box-Cox response transformations for random effect models

Almohaimeed, Amani; Einbeck, Jochen

Abstract

For the linear model with random effects of unspecified distribution, we develop methodology for simultaneous response transformation and estimation of regres- sion parameters. This is achieved by extending the Nonparametric Maximum Likelihood towards a Nonparametric Profie Maximum Likelihood technique. The methods allow to deal with overdispersion as well as two-level data scenarios.

Citation

Almohaimeed, A., & Einbeck, J. (2018, July). Box-Cox response transformations for random effect models. Presented at International Workshop on Statistical Modelling, Bristol

• Presentation Conference Type: Conference Paper (published)
• Conference Name: International Workshop on Statistical Modelling
• Acceptance Date: Apr 9, 2018
• Online Publication Date: Jul 15, 2018
• Publication Date: 2018
• Deposit Date: Jun 22, 2020
• Volume: 2
• Pages: 1-6
• Book Title: Proceedings of the 33rd International Workshop on Statistical Modelling
• Keywords: Box-Cox transformation, variance component model, EM algorithm
• Public URL: https://durham-repository.worktribe.com/output/1142503
• Publisher URL: http://www.statmod.org/workshops_archive_proceedings_2018.html
• Related Public URLs: https://www.researchgate.net/publication/327107306_Box-Cox_response_transformations_for_random_effect_models