Skip to content

Mixed models and missing p-value...

2 messages · Simon Blanchet, Christoph Buser

#
Dear R-users,

I computed a simple mixed models which was:

         mod<-lmer(nb ~ site + (1|patelle),tr)

The output was:

Linear mixed-effects model fit by REML
Formula: nb ~ site + (1 | patelle)
    Data: tr
       AIC      BIC    logLik MLdeviance REMLdeviance
  1157.437 1168.686 -574.7184   1164.523     1149.437
Random effects:
  Groups   Name        Variance Std.Dev.
  patelle  (Intercept)  34.995   5.9157
  Residual             744.736  27.2899
# of obs: 123, groups: patelle, 33

Fixed effects:
             Estimate Std. Error t value
(Intercept)  60.3483     4.3929 13.7378
siteLCN     -20.1969     7.8070 -2.5870
siteLCS     -18.2154     6.1514 -2.9612

Correlation of Fixed Effects:
         (Intr) sitLCN
siteLCN -0.563
siteLCS -0.714  0.402

I don't understand why D.F. and p-values associated to the fixed-effects 
coefficients are missing.
Could anyone help me?

When I tried another model (mod2<-lmer(nb ~ site + 
(1|patelle),tr,family=poisson)), D.F. and p-values were given...

Thank you in advance.

Very sincerely, Simon



BLANCHET Simon
PhD student
Universit?? Laval - Qu??bec-Oc??an / CIRSA
Pavillon Alexandre-Vachon
Local 8022
Qu??bec (Qu??bec), Canada G1K 7P4
T??l??phone : (418) 656-2131 poste 8022
courriel : simon.blanchet at giroq.ulaval.ca
1 day later
#
Dear Simon

There was a discussion about degrees of freedom in linear mixed
models on the R-list last week. Have a look at

https://stat.ethz.ch/pipermail/r-help/2006-January/085560.html

and following. There have been earlier discussions about that
topic. Try also:

RSiteSearch("degree freedom mixed")


Regards,

Christoph Buser

--------------------------------------------------------------
Christoph Buser <buser at stat.math.ethz.ch>
Seminar fuer Statistik, LEO C13
ETH (Federal Inst. Technology)	8092 Zurich	 SWITZERLAND
phone: x-41-44-632-4673		fax: 632-1228
http://stat.ethz.ch/~buser/
--------------------------------------------------------------


Simon Blanchet writes:
 > Dear R-users,
 > 
 > I computed a simple mixed models which was:
 > 
 >          mod<-lmer(nb ~ site + (1|patelle),tr)
 > 
 > The output was:
 > 
 > Linear mixed-effects model fit by REML
 > Formula: nb ~ site + (1 | patelle)
 >     Data: tr
 >        AIC      BIC    logLik MLdeviance REMLdeviance
 >   1157.437 1168.686 -574.7184   1164.523     1149.437
 > Random effects:
 >   Groups   Name        Variance Std.Dev.
 >   patelle  (Intercept)  34.995   5.9157
 >   Residual             744.736  27.2899
 > # of obs: 123, groups: patelle, 33
 > 
 > Fixed effects:
 >              Estimate Std. Error t value
 > (Intercept)  60.3483     4.3929 13.7378
 > siteLCN     -20.1969     7.8070 -2.5870
 > siteLCS     -18.2154     6.1514 -2.9612
 > 
 > Correlation of Fixed Effects:
 >          (Intr) sitLCN
 > siteLCN -0.563
 > siteLCS -0.714  0.402
 > 
 > I don't understand why D.F. and p-values associated to the fixed-effects 
 > coefficients are missing.
 > Could anyone help me?
 > 
 > When I tried another model (mod2<-lmer(nb ~ site + 
 > (1|patelle),tr,family=poisson)), D.F. and p-values were given...
 > 
 > Thank you in advance.
 > 
 > Very sincerely, Simon
 > 
 > 
 > 
 > BLANCHET Simon
 > PhD student
 > Universit?? Laval - Qu??bec-Oc??an / CIRSA
 > Pavillon Alexandre-Vachon
 > Local 8022
 > Qu??bec (Qu??bec), Canada G1K 7P4
 > T??l??phone : (418) 656-2131 poste 8022
 > courriel : simon.blanchet at giroq.ulaval.ca
 > 
 > ______________________________________________
 > R-help at stat.math.ethz.ch mailing list
 > https://stat.ethz.ch/mailman/listinfo/r-help
 > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html