[R-sig-ME] [EXT] Selecting a family in glmmTMB for continuous proportion data that includes 0 and 1

Daniel Lüdecke d@|uedecke @end|ng |rom uke@de
Sun Jun 4 11:29:27 CEST 2023


You can use the ordered beta family (https://doi.org/10.1017/pan.2022.20), which is implemented in glmmTMB (family = ordbeta()).

 

I also setup a page with overviews of modelling packages/functions for various outcome types:

http://htmlpreview.github.io/?https://github.com/strengejacke/mixed-models-snippets/blob/master/overview_modelling_packages.html

 

(comments welcome)

 

Daniel

 

Von: R-sig-mixed-models <r-sig-mixed-models-bounces using r-project.org> Im Auftrag von April Martinig
Gesendet: Sonntag, 4. Juni 2023 02:44
An: r-sig-mixed-models using r-project.org
Betreff: [EXT] [R-sig-ME] Selecting a family in glmmTMB for continuous proportion data that includes 0 and 1

 

Hello,

I’ve noticed several questions about fitting proportion data with glmmTMB.

Every question I’ve encountered has been focused on 0 < x < 1. However, my data included 0 and 1. I am currently using "family=beta_family(link="logit"), ziformula= ~1” after making the 0s slightly non-zero (0.00001) and the 1s slightly smaller (0.99999) - I tried the solutions under zero inflation continuous data https://bbolker.github.io/mixedmodels-misc/glmmFAQ.html#continuous-data.

However, is there a way to appropriately fit the data as is (true zeros and true ones)?

I’ve attached an image of the distribution of the data.

Any suggestions would be much appreciated!

Take care,
April

—
Dr. April Martinig, Ph.D., M.Sc.
Postdoctoral fellow
University of New South Wales
martinig.weebly.com<http://martinig.weebly.com>
“I do not accept the conventional order of things as a given."

_______________________________________________
R-sig-mixed-models using r-project.org <mailto:R-sig-mixed-models using r-project.org>  mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

--

_____________________________________________________________________

Universitätsklinikum Hamburg-Eppendorf; Körperschaft des öffentlichen Rechts; Gerichtsstand: Hamburg | www.uke.de
Vorstandsmitglieder: Prof. Dr. Christian Gerloff (Vorsitzender), Joachim Prölß, Prof. Dr. Blanche Schwappach-Pignataro, Marya Verdel
_____________________________________________________________________

SAVE PAPER - THINK BEFORE PRINTING

	[[alternative HTML version deleted]]



More information about the R-sig-mixed-models mailing list