[R] different results from lme() and lmer()

Taro Miyagawa miyagawa at live.com
Wed Jun 1 07:58:13 CEST 2011


Hello R-help,
I'm studying an example in the R book. 
The data file is available from the link below.http://www.bio.ic.ac.uk/research/mjcraw/therbook/data/fertilizer.txt
Could you explain Why the results from lme() and lmer() are different in the following case? In other examples, I can get the same results using the two functions, but not here... 
Thank you.Miya

library(lme4)library(nlme)# object dat contains the data
> summary(lme(root~fertilizer,random=~week|plant,data=dat))Linear mixed-effects model fit by REML Data: dat        AIC      BIC    logLik  171.0236 183.3863 -79.51181
Random effects: Formula: ~week | plant Structure: General positive-definite, Log-Cholesky parametrization            StdDev    Corr  (Intercept) 2.8639832 (Intr)week        0.9369412 -0.999Residual    0.4966308       
Fixed effects: root ~ fertilizer                       Value Std.Error DF   t-value p-value(Intercept)        2.799710 0.1438367 48 19.464499   0e+00fertilizercontrol -1.039383 0.2034158 10 -5.109645   5e-04 Correlation:                   (Intr)fertilizercontrol -0.707
Standardized Within-Group Residuals:       Min         Q1        Med         Q3        Max -1.9928118 -0.6586834 -0.1004301  0.6949714  2.0225381 
Number of Observations: 60Number of Groups: 12 

> lmer(root~fertilizer+(week|plant),data=dat)Linear mixed model fit by REML Formula: root ~ fertilizer + (week | plant)    Data: dat    AIC BIC logLik deviance REMLdev 174.4 187 -81.21    159.7   162.4Random effects: Groups   Name        Variance   Std.Dev.   Corr   plant    (Intercept) 4.1416e-18 2.0351e-09                 week        8.7452e-01 9.3516e-01 0.000  Residual             2.2457e-01 4.7389e-01       Number of obs: 60, groups: plant, 12
Fixed effects:                  Estimate Std. Error t value(Intercept)        -0.1847     0.2024  -0.913fertilizercontrol  -0.7612     0.2862  -2.660
Correlation of Fixed Effects:            (Intr)frtlzrcntrl -0.707
 		 	   		  


More information about the R-help mailing list