Dear R project, I am a doctoral student in Zhejiang university in China, I am using lme function in nlme package and learning the function by Package 'nlme' document. I am writing this email for some help to build a lme model. My goal was to include two non-nested random effects in the lme model. the document described how to write the random effects while I found it difficult for me to understand. My problem are as below: In these model, site and year were considered as non nested effects. I want to set the structure in lme model the same as lmer model(the 1st model below), and I used the structure "random=list(site=~1,year=~1)"(the 2nd model below) . According to the result, the lme model was obviously wrong, for the R2 and AIC were different from that of lmer model. I want to know to get the same result as lmer model, how should I set the random argument? Below is the model and results in R, ################################################################1. lmer function lmer<-lmer(data, y~x1+x2+x3+(1|site)+(1|year)) r2(lmer)##58.8%?29.7% r.squaredGLMM(lmer)##58.8%; 29.72%;AIC=462.4 ###################################################################2. lme1 lme1<-lme(data, y~x1+x2+x3,random = list(site=~1,year=~1)) r2(lme1)## NA(can't get the result) r.squaredGLMM(lme1)##92.4%; 30.89%;AIC=533 I will be appreciated for your help. Yours sincerely, Juan Liu Juan Liu PhD candidate College of Life Sciences, Zhejiang University 866 Yuhangtang Road, Hangzhou Zhejiang 310058, P.R.China 21707095 at zju.edu.cn? liujuan_1994 at outlook.com
ask for help about lme function
3 messages · Juan Liu, Kevin E. Thorpe, Bert Gunter
You might get better answers on the r-sig-ME list. The lmer() function from lme4 handles crossed and non-nested random effects quite seamlessly. I cannot comment on whether or not lme() can as well.
Kevin E. Thorpe Head of Biostatistics, Applied Health Research Centre (AHRC) Li Ka Shing Knowledge Institute of St. Michael's Assistant Professor, Dalla Lana School of Public Health University of Toronto email: kevin.thorpe at utoronto.ca Tel: 416.864.5776 Fax: 416.864.3016 > On Jul 20, 2021, at 10:32 AM, Juan Liu <21707095 at zju.edu.cn> wrote: > > EXTERNAL EMAIL: > > Dear R project, > > I am a doctoral student in Zhejiang university in China, I am using lme function in nlme package and learning the function by Package 'nlme' document. I am writing this email for some help to build a lme model. > > My goal was to include two non-nested random effects in the lme model. the document described how to write the random effects while I found it difficult for me to understand. My problem are as below: > > > > > In these model, site and year were considered as non nested effects. I want to set the structure in lme model the same as lmer model(the 1st model below), and I used the structure "random=list(site=~1,year=~1)"(the 2nd model below) . According to the result, the lme model was obviously wrong, for the R2 and AIC were different from that of lmer model. I want to know to get the same result as lmer model, how should I set the random argument? > > Below is the model and results in R, > > > > > ################################################################1. lmer function > lmer<-lmer(data, y~x1+x2+x3+(1|site)+(1|year)) > r2(lmer)##58.8%?29.7% > r.squaredGLMM(lmer)##58.8%; 29.72%;AIC=462.4 > ###################################################################2. lme1 > lme1<-lme(data, y~x1+x2+x3,random = list(site=~1,year=~1)) > r2(lme1)## NA(can't get the result) > r.squaredGLMM(lme1)##92.4%; 30.89%;AIC=533 > > > > > I will be appreciated for your help. > > > > > Yours sincerely, > > Juan Liu > > Juan Liu > > PhD candidate > > College of Life Sciences, Zhejiang University > > 866 Yuhangtang Road, Hangzhou > > Zhejiang 310058, P.R.China > > 21707095 at zju.edu.cn? liujuan_1994 at outlook.com > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
The mixed models list is r-sig-mixed-models . nlme:lme is not really designed for crossed random effects. IIRC, it's possible, but not easy. As Kevin said, lme4:lmer is really what you should use. Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Tue, Jul 20, 2021 at 9:12 AM Kevin Thorpe <kevin.thorpe at utoronto.ca> wrote:
You might get better answers on the r-sig-ME list. The lmer() function from lme4 handles crossed and non-nested random effects quite seamlessly. I cannot comment on whether or not lme() can as well. -- Kevin E. Thorpe Head of Biostatistics, Applied Health Research Centre (AHRC) Li Ka Shing Knowledge Institute of St. Michael's Assistant Professor, Dalla Lana School of Public Health University of Toronto email: kevin.thorpe at utoronto.ca Tel: 416.864.5776 Fax: 416.864.3016
On Jul 20, 2021, at 10:32 AM, Juan Liu <21707095 at zju.edu.cn> wrote: EXTERNAL EMAIL: Dear R project, I am a doctoral student in Zhejiang university in China, I am using lme
function in nlme package and learning the function by Package 'nlme' document. I am writing this email for some help to build a lme model.
My goal was to include two non-nested random effects in the lme model.
the document described how to write the random effects while I found it difficult for me to understand. My problem are as below:
In these model, site and year were considered as non nested effects. I
want to set the structure in lme model the same as lmer model(the 1st model below), and I used the structure "random=list(site=~1,year=~1)"(the 2nd model below) . According to the result, the lme model was obviously wrong, for the R2 and AIC were different from that of lmer model. I want to know to get the same result as lmer model, how should I set the random argument?
Below is the model and results in R, ################################################################1. lmer
function
lmer<-lmer(data, y~x1+x2+x3+(1|site)+(1|year)) r2(lmer)##58.8%?29.7% r.squaredGLMM(lmer)##58.8%; 29.72%;AIC=462.4 ###################################################################2.
lme1
lme1<-lme(data, y~x1+x2+x3,random = list(site=~1,year=~1))
r2(lme1)## NA(can't get the result)
r.squaredGLMM(lme1)##92.4%; 30.89%;AIC=533
I will be appreciated for your help.
Yours sincerely,
Juan Liu
Juan Liu
PhD candidate
College of Life Sciences, Zhejiang University
866 Yuhangtang Road, Hangzhou
Zhejiang 310058, P.R.China
21707095 at zju.edu.cn? liujuan_1994 at outlook.com
[[alternative HTML version deleted]]
______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. ______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.