[R-sig-ME] How to convert the formula from LME4 to NLME?

Gang Chen gangchen at mail.nih.gov
Wed May 9 23:31:07 CEST 2012


Here is my suggestion:

> summary(fm9 <- lmer(Reaction ~ 1 + Days + (1|Subject) + (0+Days|Subject), sleepstudy, REML = 0))Linear mixed model fit by maximum likelihood

Formula: Reaction ~ 1 + Days + (1 | Subject) + (0 + Days | Subject)
   Data: sleepstudy
  AIC  BIC logLik deviance REMLdev
 1762 1778   -876     1752    1744
Random effects:
 Groups   Name        Variance Std.Dev.
 Subject  (Intercept) 584.250  24.1713
 Subject  Days         33.633   5.7994
 Residual             653.116  25.5561
Number of obs: 180, groups: Subject, 18

Fixed effects:
            Estimate Std. Error t value
(Intercept)  251.405      6.708   37.48
Days          10.467      1.519    6.89

Correlation of Fixed Effects:
     (Intr)
Days -0.194

> summary(fm10 <- lme(Reaction ~ 1 + Days, random=list(Subject=pdBlocked(list(pdIdent(~1), pdSymm(~0+Days)))), sleepstudy, method='ML'))

Linear mixed-effects model fit by maximum likelihood
 Data: sleepstudy
       AIC      BIC    logLik
  1762.003 1777.968 -876.0016

Random effects:
 Composite Structure: Blocked

 Block 1: (Intercept)
 Formula: ~1 | Subject
        (Intercept)
StdDev:    24.17127

 Block 2: Days
 Formula: ~0 + Days | Subject
            Days Residual
StdDev: 5.799409 25.55613

Fixed effects: Reaction ~ 1 + Days
                Value Std.Error  DF  t-value p-value
(Intercept) 251.40510  6.745252 161 37.27142       0
Days         10.46729  1.527826 161  6.85110       0
 Correlation:
     (Intr)
Days -0.194

Standardized Within-Group Residuals:
       Min         Q1        Med         Q3        Max
-3.9535402 -0.4672676  0.0239343  0.4625157  5.1883448

Number of Observations: 180
Number of Groups: 18


On Wed, May 9, 2012 at 5:01 PM, Michael <comtech.usa at gmail.com> wrote:
> How to convert the formula from LME4 to NLME?
>
> Hi all,
>
> We have the following fomular in LME4:
>
>  > (fm9 <- lmer(Reaction ~ 1 + Days + (1|Subject) + (0+Days|Subject),
> + sleepstudy, REML = 0))
>
> How to do the same in NLME?
>
> Thanks a lot!
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models



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