[R-sig-ME] Distributional Assumption in lmer()

[Ben Bolker]

   Yes.

  The mixed models task view 
<https://cran.r-project.org/web/views/MixedModels.html> says:

Robust/heavy-tailed estimation (downweighting the importance of extreme 
observations): robustlmm, robustBLME (Bayesian robust LME), CRTgeeDR for 
the doubly robust inverse probability weighted augmented GEE estimator. 
Some packages (brms, bamlss, mgcv with family = "scat", nlmixr2) allow 
heavy-tailed response distributions such as Student-t.

   (I just added glmmTMB and GLMMadaptive to this list of packages)


On 2024-02-10 9:45 a.m., Dimitris Rizopoulos wrote:
> It’s also available in GLMMadaptive: 
> https://drizopoulos.github.io/GLMMadaptive/articles/Custom_Models.html 
> <https://drizopoulos.github.io/GLMMadaptive/articles/Custom_Models.html>
> 
> ------------------------------------------------------------------------
> *Από:* Ο χρήστης R-sig-mixed-models 
> <r-sig-mixed-models-bounces using r-project.org> εκ μέρους του χρήστη Ben 
> Bolker <bbolker using gmail.com>
> *Στάλθηκε:* Παρασκευή, Φεβρουαρίου 9, 2024 23:11
> *Προς:* r-sig-mixed-models using r-project.org <r-sig-mixed-models using r-project.org>
> *Θέμα:* Re: [R-sig-ME] Distributional Assumption in lmer()
> 
> 
> Waarschuwing: Deze e-mail is afkomstig van buiten de organisatie. Klik 
> niet op links en open geen bijlagen, tenzij u de afzender herkent en 
> weet dat de inhoud veilig is.
> Caution: This email originated from outside of the organization. Do not 
> click links or open attachments unless you recognize the sender and know 
> the content is safe.
> 
> 
> 
>     For what it's worth I think you can probably also do this in brms, if
> you want to go down the Bayesian rabbit hole ...
> 
> On 2024-02-09 5:01 p.m., Hedyeh Ahmadi wrote:
>> Thank you for the quick and informative reply.
>>
>> Best,
>>
>> Hedyeh Ahmadi, Ph.D.
>> Statistician
>> Keck School of Medicine
>> Department of Preventive Medicine
>> University of Southern California
>>
>> LinkedIn
>> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.linkedin.com%2Fin%2Fhedyeh-ahmadi&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044801031%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=BflyM0EjMbj50sgQGJ%2FhucVgmaEwHNvTKeBcjExX6rc%3D&reserved=0<https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.linkedin.com%2Fin%2Fhedyeh-ahmadi&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044810000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=UWfXvLGMF6N20VDCaxX1TobXCRjBr6RiPTtDyQmyAwk%3D&reserved=0> <http://www.linkedin.com/in/hedyeh-ahmadi>
>>
>>
>>
>>
>> ________________________________
>> From: R-sig-mixed-models <r-sig-mixed-models-bounces using r-project.org> on behalf of Ben Bolker <bolkerb using mcmaster.ca>
>> Sent: Friday, February 9, 2024 1:34 PM
>> To: r-sig-mixed-models using r-project.org <r-sig-mixed-models using r-project.org>
>> Subject: Re: [R-sig-ME] Distributional Assumption in lmer()
>>
>>     No, but:
>>
>> (1) glmmTMB has this (family = t_family)
>> (2) you can achieve a similar goal with the robustlmm package
>>
>>     cheers
>>      Ben Bolker
>>
>>
>> On 2024-02-09 3:42 p.m., Hedyeh Ahmadi wrote:
>>> Hello All,
>>> I was wondering if there is a way to implement t-distribution assumption instead of family ="Gaussian" assumption in the lmer() function.
>>>
>>> I am asking since I have been seeing heavy tails in my outcomes and it shows up in my residual diagnostic QQplot hence I think a t-distribution would be more appropriate compared to Normal distribution.
>>>
>>> Any help would be greatly appreciated.
>>>
>>> Best,
>>>
>>> Hedyeh Ahmadi, Ph.D.
>>> Statistician
>>> Keck School of Medicine
>>> Department of Preventive Medicine
>>> University of Southern California
>>>
>>> LinkedIn
>>> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense.com%2Fv3%2F__http%3A%2F%2Fwww.linkedin.com%2Fin%2Fhedyeh-ahmadi__%3B!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGdIIuWgI%24&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044815898%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=ygMjmU88ZBOHB6GHbBzmvkOaERIvZz4oisAmezhG06A%3D&reserved=0 <https://urldefense.com/v3/__http://www.linkedin.com/in/hedyeh-ahmadi__;!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGdIIuWgI$> <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense.com%2Fv3%2F__http%3A%2F%2Fwww.linkedin.com%2Fin%2Fhedyeh-ahmadi__%3B!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGdIIuWgI%24&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044821234%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=FMIbPehAFvXEAYBcP0BcGneyPDg0d41AOF%2B4FBu%2FzqQ%3D&reserved=0 >
>>>
>>>
>>>
>>>
>>>
>>>         [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> R-sig-mixed-models using r-project.org mailing list
>>> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense.com%2Fv3%2F__https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-mixed-models__%3B!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGRRULy-w%24&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044825917%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=XXOQQb3Bq2TsD15mA4JLybKm%2FfcunRxP5h4LqfPjibg%3D&reserved=0 <https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models__;!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGRRULy-w$>
>>
>> --
>> Dr. Benjamin Bolker
>> Professor, Mathematics & Statistics and Biology, McMaster University
>> Director, School of Computational Science and Engineering
>>    Associate chair (graduate, math), Mathematics & Statistics
>>     > E-mail is sent at my convenience; I don't expect replies outside of
>> working hours.
>>
>> _______________________________________________
>> R-sig-mixed-models using r-project.org mailing list
>> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense.com%2Fv3%2F__https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-mixed-models__%3B!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGRRULy-w%24&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044830659%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=%2BmYtfJH%2BgN0%2F%2F7BU%2BVPYkAJeen7r%2B8XAPHnLttJdywk%3D&reserved=0 <https://urldefense.com/v3/__https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models__;!!LIr3w8kk_Xxm!owK3rp5BYaR9BZh-ZgTlsXIxdM3aebgcexnD7meyFOmfUi7AnsDslFD8XZYGRhYRJxXflAbk6pYGRRULy-w$>
>>
>>       [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-mixed-models using r-project.org mailing list
>> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-mixed-models&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044835348%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=t4c9TyhKRgr31Q2ENbFuv8gcHZK8wAAfhEtOLGCBG%2FM%3D&reserved=0 <https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models>
> 
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-mixed-models&data=05%7C02%7Cd.rizopoulos%40erasmusmc.nl%7Caebf617ef97a430fd59808dc29bc1797%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C638431135044840607%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=wsMypblJAp8STHQ28siavAUI5yMVM9JCEEGssnpc2PA%3D&reserved=0 <https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models>