[R-sig-ME] random effects estimation in Lmer

[Henric (Nilsson) Winell]

On 2009-12-15 09:23, David Duffy wrote:

> On Tue, 15 Dec 2009, Hester Lingsma wrote:
> 
>> Dear R-users,
>> I have a question on how the random effects are estimated in Lmer
>>
>> I have a dataset with families with women who have breastcancer. The age
>> of cancer diagnosis is highly variable between the families. We want to
>> predict the age of cancer diagnosis (for screening purposes) for 'new'
>> family members. I have fitted a linear random effects model with lmer 
>> (model
>> 1)  with age at diagnosis as outcome, a random intercept for family, and
>> cancer type (three subtypes, on family level) in the fixed part.
>>
> 
> Have you looked at the kinship package at all?  It uses lme, but 
> analyses age at onset data from families more nicely than lmer can (see 
> the coxme() function and the vignette/documentation).

Actually, the latest version of the coxme() function now lives in Terry 
Therneau's new 'coxme' package.

It no longer depends on 'nlme', but uses its own fitting routines and 
sparse matrix representation, and has a lme4-like syntax for the 
specification of random effects.


Henric



> 
> David Duffy.
>