be trusted with s(...,bs="re") terms, and can be poor if the
> models being compared involve any terms that can be penalized to zero
> during fitting. (Although the mechanisms are a little different, this is
> similar to the problem we would have if the models were viewed as
97.10 169.00 434.0 <2e-16
#s(Word,YearRec.z) 128.98 170.00 867.4 <2e-16
#
#R-sq.(adj) = 0.372 Deviance explained = 32.4%
#REML score = 97449 Scale est. = 1 n = 69259
Met vriendelijke groet,
Martijn
--
*
n
1.7-17: summary(model) followed by summary(model, freq=F). Data can be
sent offline if required.
With kind regards,
Martijn Wieling,
University of Groningen
# call in all versions
model = bam(NoStandard3 ~
te(GeoX,GeoY,WordFreqLog.z,by=OldSpeakerContrast,d=c(2,1)) +
OldSpeakerContrast + PopCnt
uot;), data=wrddst, method="REML")
summary(modelBAMnew)
summary(modelBAMnew,freq=F) # same results as modelBAMold
# using version 1.7-17, default: fREML - contains a bug
modelBAMfREML <- bam(RefPMIdistMeanLog.c ~ SpYearBirth.z*IsAragon +
SpIsMale + s(Key,bs="re&qu
a great deal - see also below).
With kind regards,
Martijn Wieling
University of Groningen
lme4 model (most recent version of lme4)
modelLMER <- lmer(RefPMIdistMeanLog.c ~ SpYearBirth.z*IsAragon +
SpIsMale + (1|Key), data=wrddst)
#Estimate Std. Error t value
#SpYea
mary of
the model using method="fREML" is also shown below.
I hope someone can shed some light on this.
With kind regards,
Martijn Wieling,
University of Groningen
#
### mgcv version 1.7-11
#
Family: gaussian
Link function
d really appreciate it if someone could help shed light on these
findings... Perhaps I'm doing something wrong, but the results go
against my intuition.
With kind regards,
Martijn Wieling,
University of Groningen
### breakpoint analysis using lm, both with and without covariate
# note that AoEO
d-effect factor).
With kind regards,
Martijn
--
*******
Martijn Wieling
http://www.martijnwieling.nl
wiel...@gmail.com
+31(0)614108622
***
University of Groningen
http://www.rug.nl/staff/m.b.wieling
***
On Fri, May 11, 2012
the interaction, might
result in an actual decrease of the degrees of freedom invested by the
total model (caused by a decrease of the edf's of splines in the model
with the interaction). In this case, how would I proceed in
determining if the model including the interaction term is be
Wieling
http://www.martijnwieling.nl
wiel...@gmail.com
+31(0)614108622
***
University of Groningen
http://www.rug.nl/staff/m.b.wieling
***
On Tue, Apr 24, 2012 at 11:22, Martijn Wieling wrote:
> Hi Simon,
>
&g
ould lead to unexpected constraints being applied to s(...,bs="re")
> terms in some cases.
>
> So, could you sent me the gam call that generates the problem, and
> perhaps try out if it still happens in 1.7-13?
>
> best,
> Simon
>
> On 23/04/12 18:26, Martijn Wieli
y-word random slopes
than words, and/or ii) how I can link the random slopes which are
present to the correct labels of the words?
(I did not include the model as it is >300 MB in size, but let me know
if this is necessary.)
Any help would be greatly appreciated!
With kind regards,
Martijn Wie
e error. I've tried to locate the error and it appears to
occur in the function pinvXVX in the block which 'deals with the
fractional part of the pinv'.
Any help would be appreciated!
With kind regards,
Martijn Wieling
University of Groningen
http://www.martijnwieling.nl
_
13 matches
Mail list logo
