[R-sig-ME] Simulating Data from a Generalized Linear Mixed Model

[Afshartous, David]

Chuck,
Gelman's book Data Analysis Using Regression and Multilevel/Hierarchical
Model has
good examples in both R and Bugs; see p.363 and also look in the index
under "fake data simulation" for other examples.
Cheers,
David
 

> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org 
> [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf 
> Of Chuck Cleland
> Sent: Friday, August 24, 2007 8:42 AM
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] Simulating Data from a Generalized Linear 
> Mixed Model
> 
> Hello:
>   Does anyone have suggestions for how one can simulate data 
> for a generalized linear mixed model?  For example, below are 
> summaries of two models fit by lmer() for which I would like 
> to simulate new data.  If the summaries do not provide 
> sufficient information to generate new observations, what 
> other information is needed?
> 
> thanks for any help,
> 
> Chuck
> 
> > summary(fm1)
> Linear mixed-effects model fit by REML
> Formula: Y ~ V7 * TIME + (TIME | id)
>    Data: df.uni
>   AIC  BIC logLik MLdeviance REMLdeviance
>  8795 8836  -4390       8763         8781
> Random effects:
>  Groups   Name Variance Std.Dev. Corr
>  id            0.23671  0.48653
>                0.52332  0.72341  -0.079
>  Residual      0.51130  0.71506
> number of obs: 2611, groups: id, 480
> 
> Fixed effects:
>             Estimate Std. Error t value
> (Intercept)  0.02499    0.04311   0.580
> V7          -0.08775    0.06027  -1.456
> TIME         0.15238    0.04826   3.158
> V7:TIME      0.29308    0.06717   4.364
> 
> Correlation of Fixed Effects:
>         (Intr) V7     TIME
> V7      -0.715
> TIME    -0.106  0.076
> V7:TIME  0.076 -0.107 -0.718
> 
> > summary(fm2)
> Generalized linear mixed model fit using Laplace
> Formula: Y ~ V7 * TIME + (TIME | id)
>    Data: df.uni
>  Family: binomial(logit link)
>   AIC  BIC logLik deviance
>  2278 2319  -1132     2264
> Random effects:
>  Groups Name Variance Std.Dev. Corr
>  id          0.63468  0.79667
>              2.96036  1.72057  0.125
> number of obs: 2601, groups: id, 480
> 
> Estimated scale (compare to  1 )  0.6274443
> 
> Fixed effects:
>             Estimate Std. Error z value Pr(>|z|)
> (Intercept)  0.29549    0.12484   2.367   0.0179 *
> V7           0.08769    0.17198   0.510   0.6101
> TIME        -0.67834    0.14820  -4.577 4.71e-06 ***
> V7:TIME      0.45752    0.20165   2.269   0.0233 *
> ---
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> 
> Correlation of Fixed Effects:
>         (Intr) V7     TIME
> V7      -0.726
> TIME    -0.260  0.189
> V7:TIME  0.191 -0.253 -0.735
> 
> --
> Chuck Cleland, Ph.D.
> NDRI, Inc.
> 71 West 23rd Street, 8th floor
> New York, NY 10010
> tel: (212) 845-4495 (Tu, Th)
> tel: (732) 512-0171 (M, W, F)
> fax: (917) 438-0894
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>