Skip to content

Clarification needed about the terminology used when talking about mixed models

4 messages · Thompson,Paul, Amelie LESCROEL, Paolo Innocenti

Original
Thompson,Paul
# 
I can start.

1) Mixed models are models which include fixed and random effects.  Mixed models can be used to estimate and test multi-level models, and random regression models are among those.  Random regression models are a form of mixed model, in which each unit at Level 1 has an individualized regression equation estimated.  The regression often includes a slope (individual rate of change) and intercept (individually tailored y-intercept).

2) I have no idea what an animal model is, unless you mean one of those little dinosaur toys :-).

-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org on behalf of Amelie Lescroel
Sent: Mon 8/9/2010 7:27 AM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] Clarification needed about the terminology used whentalking about mixed models
 
Hello,

 

Could someone give me a clear definition of what a "random regression model"
is and what are the differences between a mixed model, a random regression
model and an animal model? I'm beginning to be lost among the multiple
designations. 

 

Thanks,

 

Amelie

 

 

 



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


-----------------------------------------------------------------------
Confidentiality Notice: This e-mail message, including any attachments,
is for the sole use of the intended recipient(s) and may contain
privileged and confidential information.  Any unauthorized review, use,
disclosure or distribution is prohibited.  If you are not the intended
recipient, please contact the sender by reply e-mail and destroy
all copies of the original message.
Amelie LESCROEL
# 
Thanks a lot Paul. Thus I should be able to fit a random regression model
using lmer. You'll probably hear from me at that time!

-----Original Message-----
From: Thompson,Paul [mailto:Paul.Thompson at sanfordhealth.org] 
Sent: Monday, August 09, 2010 3:08 PM
To: Amelie Lescroel; r-sig-mixed-models at r-project.org
Subject: RE: [R-sig-ME] Clarification needed about the terminology used
whentalking about mixed models


I can start.

1) Mixed models are models which include fixed and random effects.  Mixed
models can be used to estimate and test multi-level models, and random
regression models are among those.  Random regression models are a form of
mixed model, in which each unit at Level 1 has an individualized regression
equation estimated.  The regression often includes a slope (individual rate
of change) and intercept (individually tailored y-intercept).

2) I have no idea what an animal model is, unless you mean one of those
little dinosaur toys :-).

-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org on behalf of Amelie Lescroel
Sent: Mon 8/9/2010 7:27 AM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] Clarification needed about the terminology used
whentalking about mixed models
 
Hello,

 

Could someone give me a clear definition of what a "random regression model"
is and what are the differences between a mixed model, a random regression
model and an animal model? I'm beginning to be lost among the multiple
designations. 

 

Thanks,

 

Amelie

 

 

 



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


-----------------------------------------------------------------------
Confidentiality Notice: This e-mail message, including a...{{dropped:7}}
Paolo Innocenti
# 
An animal model is a mixed model in which you use a "pedigree" to 
estimate your variance components, instead of defined levels within a 
factor. This pedigree provides information about the relationships among 
your observations. I have no idea about how it works, but Jarrod 
Hadfield provides this reference when talking about animal models in his 
MCMCglmm package (in which you can fit animal models):

C. R. Henderson. Simple method for computing inverse of a numerator 
relationship matrix used in prediction of breeding values. Biometrics, 
32(1):69?83,
1976.

Someone should correct me if what I said is wrong =).
Best,
paolo
Thompson,Paul wrote: