[R-sig-ME] Residual variance for mixed effects survival analysis

Farrar, David Farrar.David at epa.gov
Tue Aug 18 21:30:52 CEST 2015

It would seem odd to me if there have been no models for this situation with analogues of extra-binomial variation and some sort of time series structure, although such a model might not be termed "Cox" modeling.


-----Original Message-----
From: R-sig-mixed-models [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Bradley Carlson
Sent: Tuesday, August 18, 2015 12:39 PM
To: Ken Beath
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Residual variance for mixed effects survival analysis

I feared that might be the situation - thanks for confirming. I'll be toying with some creative options then. Thanks again,


On Mon, Aug 17, 2015 at 7:26 PM, Ken Beath <ken.beath at mq.edu.au> wrote:

> For Cox survival analysis there is no residual as part of the model, 
> hence no residual variance.
> Ken
> On 18 August 2015 at 02:01, Bradley Carlson <carbrae at gmail.com> wrote:
>> I'm working on an analysis of the amount of time it takes an animal 
>> to perform a behavior. This data is right-censored, as we stopped 
>> waiting after 10 minutes and simply had to record "10+ minutes" for 
>> any animals that didn't perform the behavior. The data is well suited 
>> for a Cox proportional hazards 'survival' analysis in this regard. 
>> However, I have multiple measurements of each individual animal 
>> (random effect) and would like to quantify the individual 
>> repeatability of the behavior in addition to the effect of 
>> covariates. In a typical LMM, this would be the (variance among 
>> random intercepts) / (variance among random intercepts + residual variance).
>> I can fit a mixed effects survival analysis in the package 'coxme'. 
>> It provides a random effect variance, but no residual variance. Is it 
>> possible, given the mechanics of fitting an ME survival analysis, to 
>> get a residual variance?
>> Thank you in advance!
>> --
>> Bradley Evan Carlson
>> Assistant Professor of Biology
>> Wabash College, Crawfordsville IN
>> Email: *carlsonb at wabash.edu* <+carlsonb at wabash.edu>
>> Website: https://sites.google.com/site/bradleyecarlson/home
>>         [[alternative HTML version deleted]]
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list 
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> --
> *Ken Beath*
> Lecturer
> Statistics Department
> Phone: +61 (0)2 9850 8516
> Level 2, AHH
> http://stat.mq.edu.au/our_staff/staff_-_alphabetical/staff/beath,_ken/
> CRICOS Provider No 00002J
> This message is intended for the addressee named and=2...{{dropped:7}}

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