LISTSERV mailing list manager LISTSERV 16.0

Help for R-USERS-L Archives


R-USERS-L Archives

R-USERS-L Archives


R-USERS-L@LISTS.UFL.EDU


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

R-USERS-L Home

R-USERS-L Home

R-USERS-L  2018

R-USERS-L 2018

Subject:

Re: Mixed effect models

From:

"Chaudhary,Vratika" <[log in to unmask]>

Reply-To:

UF R Users List <[log in to unmask]>

Date:

Tue, 17 Jul 2018 19:47:46 +0000

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (37 lines)

Thank you for the reply. I have a dataset where a ‘success’ (good ) or ‘failure’ (bad) of a series of questions for individual datapoint (publication) depend on the conservation status of the study species, year of publication and the kind of model used amongst other things. On performing logistic regression using only the fixed effects, one of the concern of reviewers was if  we looked into the effect of the same species and same authors in the model. So I chose mixed effect models to account for pseudo replications in the data.

And on performing model selection, there isn’t a whole lot difference between models with fixed and random effect. And I couldn’t find a whole lot of resources to make that selection, or to perform model averaging between the effects. I will appreciate any input on this and please let me know if I can explain the contact in more details.

thank you
Vratika

> On 17-Jul-2018, at 3:17 PM, Ben Bolker <[log in to unmask]> wrote:
> 
>  I don't. I have to say I'm not even a fan of the model-*testing*
> approach to choosing between fixed and random effects (i.e. the
> Hausman test, mostly used in econometrics); I would prefer to choose
> fixed vs random for a particular effect on first principles (or on
> pragmatic grounds, e.g. I don't have enough levels/the model is
> singular).
>   Can you give more context/say more about your goals?
> On Tue, Jul 17, 2018 at 2:34 PM Chaudhary,Vratika <[log in to unmask]> wrote:
>> 
>> Hi everyone
>> 
>> It's less of R , and more of stats question. Does anyone have links/resources/literature on model averaging between fixed and random effects logistic regression models.
>> Or know if model averaging between these two groups is a correct approach?
>> 
>> Thank you
>> Vratika
>> 
>> This list strives to be beginner friendly.  However, we still ask that you
>> PLEASE do read the posting guide https://urldefense.proofpoint.com/v2/url?u=http-3A__www.R-2Dproject.org_posting-2Dguide.html&d=DwIDAg&c=pZJPUDQ3SB9JplYbifm4nt2lEVG5pWx2KikqINpWlZM&r=3MJaagrtDUL79e1ACXZji8lYXjs07_ujkEIg1it6gQ0&m=yQpUGW7CoTvv2GqLd59yrYKI1wUvzkK2RB7EsMQxQ10&s=_IJEmrcG2xfBR0I9o76-8U9ZDmC4Kz4NP80Vnq64Ptk&e=
>> and provide commented, minimal, self-contained, reproducible code.
> 
> This list strives to be beginner friendly.  However, we still ask that you
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

This list strives to be beginner friendly.  However, we still ask that you
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Top of Message | Previous Page | Permalink

Advanced Options


Options

Log In

Log In

Get Password

Get Password


Search Archives

Search Archives


Subscribe or Unsubscribe

Subscribe or Unsubscribe


Archives

2019
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
2008

ATOM RSS1 RSS2



LISTS.UFL.EDU

CataList Email List Search Powered by the LISTSERV Email List Manager