Can we do b-spline within lmer function?
Dear Zhong, Two (probably) minor comments: (i) As stated in ?bs, "bs(x)" defaults to a simple third degree polynomial, therefore if you wants real splines you have to play with the argument df and/or knots. (ii) In theory you can move to penalized B-splines (i.e., P-splines which appear to be preferable to simple b-splines) in a mixed model framework. However lmer and friends may not be used because it does not appear possible to specify the proper covariance matrix for the random effects (you should use a multiple of identity matrix as pdDiag in nlme::lme()) all the best, vito Zhong, Xiao ha scritto:
Thanks, Ken.
I tried your advice on my model:
model2.growth.mcas5 <- lmer(response ~ bs(monthElapsed) + skills + (1|studentID), data= mcas5, family=binomial(link="logit"), control =
list(msVerbose = 1, usePQL = FALSE))
It worked! What it gave me is:
(Intercept) bs(monthElapsed)1 bs(monthElapsed)2 bs(monthElapsed)3
-0.1509712 0.7019262 -0.1934738 0.6257330
skillsG-Geometry skillsM-Measurement skillsN-Number-Sense-Operations skillsP-Patterns-Relations-Algebra
-0.4333880 -0.9468761 0.5263370 0.1897903
Since I am trying to use design matrix to predict after that glmm fitting, could you tell me if there is some function/command that I can use to find what the three "bs" did to my "monthElapsed" variable?
Thanks!
Xiao
________________________________________ From: Ken Beath [kjbeath at kagi.com] Sent: Monday, January 14, 2008 3:29 PM To: Zhong, Xiao Cc: r-sig-mixed-models at r-project.org Subject: Re: [R-sig-ME] Can we do b-spline within lmer function? On 15/01/2008, at 2:59 AM, Zhong, Xiao wrote: Hello, I am working on a project that uses glmm. I would like to consider b- spline in the original glmm but I couldn't find how to add b-spline terms to the normal lmer function. Is there anybody who could help me with that? library(splines) fm2 <- lmer(Reaction ~ bs(Days) + (1|Subject), sleepstudy) Ken _______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
==================================== Vito M.R. Muggeo Dip.to Sc Statist e Matem `Vianelli' Universit? di Palermo viale delle Scienze, edificio 13 90128 Palermo - ITALY tel: 091 6626240 fax: 091 485726/485612