Has the latest version of glmmTMB remove the ability to fix variance components in a mixed model? i.e. https://stackoverflow.com/questions/77585084/fix-variance-components-in-a-mixed-model-in-r-glmmtmb-or-sommer
Fixing variance components in glmmTMB
4 messages · Tempelman, Robert, Ben Bolker
Not intentionally/not that I know of. Is this approach not working
for you? Can you provide a reproducible example of what you want to do
and why it's not working, either here or on Stack Overflow or in the
glmmTMB issues list?
cheers
Ben Bolker
On 2025-01-21 2:33 p.m., Tempelman, Robert wrote:
Has the latest version of glmmTMB remove the ability to fix variance components in a mixed model? i.e. https://stackoverflow.com/questions/77585084/fix-variance-components-in-a-mixed-model-in-r-glmmtmb-or-sommer [[alternative HTML version deleted]]
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
Got it working now...somebody passed onto me this nugget last night: As of version 1.1.9<https://urldefense.com/v3/__https://cloud.r-project.org/web/packages/glmmTMB/news.html__;!!HXCxUKc!3__nPeuzixezIKH5upUs1PNx5jNLFsdcOqcwQWHN51gD4oFrwbgDeYztNRxll4gyEb7taAg9_BAvDR22KK2sHOPh$>, the argument corresponding to the residual variance (?betad?) is parameterized on the log-SD scale instead of the log-variance scale. In version 1.1.10, this argument changed from ?betad? to ?betadisp?. I had "betad" in all of my older code.
From: R-sig-mixed-models <r-sig-mixed-models-bounces at r-project.org> on behalf of Ben Bolker <bbolker at gmail.com>
Sent: Wednesday, January 22, 2025 10:09 AM
To: r-sig-mixed-models at r-project.org <r-sig-mixed-models at r-project.org>
Subject: Re: [R-sig-ME] Fixing variance components in glmmTMB
Sent: Wednesday, January 22, 2025 10:09 AM
To: r-sig-mixed-models at r-project.org <r-sig-mixed-models at r-project.org>
Subject: Re: [R-sig-ME] Fixing variance components in glmmTMB
Not intentionally/not that I know of. Is this approach not working
for you? Can you provide a reproducible example of what you want to do
and why it's not working, either here or on Stack Overflow or in the
glmmTMB issues list?
cheers
Ben Bolker
On 2025-01-21 2:33 p.m., Tempelman, Robert wrote:
> Has the latest version of glmmTMB remove the ability to fix variance components in a mixed model? i.e. https://urldefense.com/v3/__https://stackoverflow.com/questions/77585084/fix-variance-components-in-a-mixed-model-in-r-glmmtmb-or-sommer__;!!HXCxUKc!0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o-UAvt5UX4EQ3yV-1PGenM$
>
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models__;!!HXCxUKc!0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o-UAvt5UX4EQ3yVpoWilJo$
_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models__;!!HXCxUKc!0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o-UAvt5UX4EQ3yVpoWilJo$
OK. As you point out, both of these are in the NEWS: the change in variance scale is admittedly something you just have to know/read about. The change from 'betad' to 'betadisp' is something that it might be possible to flag with an informative error message ...
On 2025-01-22 10:14 a.m., Tempelman, Robert wrote:
Got it working now...somebody passed onto me this nugget last night: As of version 1.1.9 <https://urldefense.com/v3/__https://cloud.r- project.org/web/packages/glmmTMB/news.html__;!!HXCxUKc! 3__nPeuzixezIKH5upUs1PNx5jNLFsdcOqcwQWHN51gD4oFrwbgDeYztNRxll4gyEb7taAg9_BAvDR22KK2sHOPh$>, the argument corresponding to the residual variance (?betad?) is parameterized on the log-SD scale instead of the log-variance scale. ??In version 1.1.10, this argument changed from ?betad? to ?betadisp?. I had "betad" in all of my older code. ------------------------------------------------------------------------ *From:* R-sig-mixed-models <r-sig-mixed-models-bounces at r-project.org> on behalf of Ben Bolker <bbolker at gmail.com> *Sent:* Wednesday, January 22, 2025 10:09 AM *To:* r-sig-mixed-models at r-project.org <r-sig-mixed-models at r-project.org> *Subject:* Re: [R-sig-ME] Fixing variance components in glmmTMB ?? Not intentionally/not that I know of.? Is this approach not working for you?? Can you provide a reproducible example of what you want to do and why it's not working, either here or on Stack Overflow or in the glmmTMB issues list? ?? cheers ??? Ben Bolker On 2025-01-21 2:33 p.m., Tempelman, Robert wrote:
Has the latest version of glmmTMB remove the ability to fix variance components in a mixed model?? i.e. https://urldefense.com/v3/__https://stackoverflow.com/
questions/77585084/fix-variance-components-in-a-mixed-model-in-r- glmmtmb-or-sommer__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yV-1PGenM$ <https://urldefense.com/v3/__https:// stackoverflow.com/questions/77585084/fix-variance-components-in-a-mixed- model-in-r-glmmtmb-or-sommer__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yV-1PGenM$>
??????? [[alternative HTML version deleted]]
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig-
mixed-models__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yVpoWilJo$ <https://urldefense.com/v3/__https://stat.ethz.ch/ mailman/listinfo/r-sig-mixed-models__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yVpoWilJo$>
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig- mixed-models__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yVpoWilJo$ <https://urldefense.com/v3/__https://stat.ethz.ch/ mailman/listinfo/r-sig-mixed-models__;!!HXCxUKc! 0JqqzBv9yhXtOwmzXwSK3-59lOHI4g6zhLf-6JJ68JqDL9s1FxMJj8YE8o- UAvt5UX4EQ3yVpoWilJo$>
Dr. Benjamin Bolker Professor, Mathematics & Statistics and Biology, McMaster University Director, School of Computational Science and Engineering > E-mail is sent at my convenience; I don't expect replies outside of working hours.