Hi Zhao,
This is not a direct answer to your question, but a suggestion for a
different approach. The ordinal package was designed to cope with
issues like this (parameter constraints in ordinal regression models)
- try the following:
> library(ordinal)
> data(wine, package="ordinal")
> ## Fit mo
A third, and often preferable, way is to add an observation-level random effect:
library(lme4)
data1$obs <- factor(seq_len(nrow(data1)))
model <- glmer(y ~ x1 + x2 + (1 | obs), family=poisson(link=log), data=data1)
See http://glmm.wikidot.com/faq and search for "individual-level
random effects".
On 5 January 2015 at 21:08, Ben Bolker wrote:
> Roger Coppock cox.net> writes:
>
>>
>> When will "R" implement the "se.fit" option to the
>> predict.nls() function? Is there some schedule?
>>
>
> I think this is unlikely to happen, ever (sorry). The exact method
> for finding confidence inte
On 26 November 2014 at 17:55, Charlotte Whitham
wrote:
> Dear Rune,
>
> Thank you for your prompt reply and it looks like the ordinal package could
> be the answer I was looking for!
>
> If you don't mind, I'd also like to know please what to do if the tests show
> the proportional odds assumpti
Dear Charlie,
I admit that I haven't read your email closely, but here is a way to
test for non-proportional odds using the ordinal package (warning:
self-promotion) using the wine data set also from the ordinal package.
There is more information in the package vignettes
Hope this is something yo
Aurore,
I don't know if car::Anova is able/should be able to produce anova
tables for clmm objects; I usually use drop1() (and sometimes add1) to
test terms in CLMMS:
> library(ordinal)
> fm1 <- clmm(rating ~ temp + contact + (1|judge), data=wine)
> drop1(fm1, test="Chi")
Single term deletions
M
It's telling you that one or more of the grouping factors for the
random-effect terms has less than three levels. From what you write,
this seems to apply to Location: you may want to treat it as a
fixed-effect instead.
Hope this helps,
Rune
On 2 June 2014 14:00, adesgroux wrote:
> Dear all,
>
>
Dear Caroline,
Yes, it seems you have complete separation for the 'Timepoint'
variable. This means that the likelihood is unbounded for that
parameter and the optimizer just terminates when it gets far enough
out on an asymptote and improvements are below a threshold. This is
also the reason the v
Yes; see clm and clmm2 (mixed effects) in the ordinal package for
fitting proportional odds models. See section 3 of
http://cran.r-project.org/web/packages/ordinal/vignettes/clm_tutorial.pdf
to see how to test the proportional odds assumption with clm - it is
equivalent for clmm2 models. For an int
There is no argument 'test' to anova.clm hence the error message.
The likelihood ratio statistic (or, alternatively, G^2 statistic or
Deviance statistic) has an asymptotic chi-square distribution, so it
is the size of that statistic your reviewers are asking for. It is
printed in the anova output
Try
library(lmerTest)
fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
summary(fm1)
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
Data: sleepstudy
AIC BIC logLik deviance REMLdev
1756 1775 -871.8 17521744
Random effects:
Groups NameV
rmat. I would appreciate that if you could point me to
> the right direction. Also, I know I am dealing with a relatively large
> data set, but is there any way to speed up the estimation a bit.
> Thanks a lot!
>
> Jun
>
> On Fri, Jun 7, 2013 at 1:04 AM, Rune Haubo wrote:
On 6 June 2013 00:13, Xu Jun wrote:
> Dear r-helpers,
>
> I have two questions on multilevel binary and ordered regression models,
> respectively:
>
> 1. Is there any r function (like lmer or glmer) to run multilevel ordered
> regression models?
Yes, package ordinal will fit such models.
Cheers,
On 18 April 2013 18:38, Thomas Foxley wrote:
> Rune,
>
> Thank you very much for your response.
>
> I don't actually have the models that failed to converge from the first
> (glmulti) part as they were not saved with the confidence set. glmulti
> generates thousands of models so it seems reasonabl
On 15 April 2013 13:18, Thomas wrote:
>
> Dear List,
>
> I am using both the clm() and clmm() functions from the R package 'ordinal'.
>
> I am fitting an ordinal dependent variable with 5 categories to 9 continuous
> predictors, all of which have been normalised (mean subtracted then divided
> b
ou can also make similar tests with the VGAM package, but I
am not as well versed in that package.
Hope this helps,
Rune
Rune Haubo Bojesen Christensen
Postdoc
DTU Compute - Section for Statistics
---
Technical University of Denmark
Department of
Hi Alice,
A factor is a fairly basic R concept that you can read about in
http://cran.r-project.org/doc/manuals/R-intro.pdf on page 16. Now to
fit the CLM, you need to turn your response variable into a factor
with something like
datareg$Newpercentagecash <- factor(datareg$Newpercentagecash, orde
Den 11/09/2012 16.36 skrev "Anera Salucci" :
>
> Hi all,
>
> I am trying to fit a random effect model to categorical response
variable using package "ordinal" /"clmm".
>
> How can I find the correlation between random effects (random intercept
and random slope)
You cannot, as such models are not
ddress this error, I would very much
> appreciate your response.
>
> Thank you in advance.
>
> Jeremy
>
> Date File Attachment (200 rows):
> http://r.789695.n4.nabble.com/file/n4635829/20120709_JLittle_data_file.txt
> 20120709_JLittle_data_file.txt
>
>
> --
> View this message in context
25/06/2012 09:32, Rune Haubo wrote:
>>
>> According to standard likelihood theory these are actually not
>> t-values, but z-values, i.e., they asymptotically follow a standard
>> normal distribution under the null hypothesis. This means that you
>
>
> Whose 'st
t; Website:
> http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm
>
> [[alternative HTML version deleted]]
>
>
> __
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mail
lt University
> --
> View this message in context:
> http://r.789695.n4.nabble.com/Ordinal-logistic-regression-p-values-tp3777674p368.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______
> R-help@r-project.org ma
://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.
>
--
Rune Haubo Bojesen Christensen
Ph.D. Student, M.Sc. Eng.
Phone: (+45) 45 25 33 63
Mobile:
lmer is not designed for ordered categorical data as yours are. You could
take a look at the ordinal package which is designed for this type of data
including mixed models (function clmm) which you probably want to use.
Best,
Rune
Den 24/03/2011 21.03 skrev "Rasanga Ruwanthi" :
>
> Dear List,
>
>
.
- helpful package vignettes.
- implementation of core functions in C.
Comments, critique, suggestions, wishes and contributions are always
highly appreciated.
Kind regards
Rune
--
Rune Haubo Bojesen Christensen
PhD student, M.Sc. Eng.
Phone: (+45) 45 25 33 63
Mail: rhbc at imm.dtu.dk
lihood, so they should be, I think.
>> Any help would be appreciated.
>>
>> Corey
>>
>> Corey S. Sparks, Ph.D.
>>
>> Assistant Professor
>> Department of Demography and Organization Studies
>> University of Texas San Antonio
>> One UTSA Circle
>&g
el:
summary(glm(grade > 3 ~ score, family = binomial))
as well as by lrm in package Design.
>
> Of course profile-Lik based CI may be very usefuls at this aim..but this is
> another story..
I agree, but the topic is closely related to standard errors ;-)
Best
Rune
>
> many tha
Dear Vito
No, you are not wrong, but you should center score prior to model estimation:
summary(fm1 <- polr(factor(grade)~I(score - mean(score
which gives the same standard errors as do lrm. Now the intercepts
refer the median score rather than some potential unrealistic score of
0.
You can
Hi Richard
You are trying to compare two models, that are not nested. This means
that all usual asymptotics of the test statistics break down, hence
the (second) test you are attempting is not meaningful. Usually one
decides on the form of the response on other grounds such as residual
analysis or
Hi Stephen
On 22/02/2008, Stephen Cole <[EMAIL PROTECTED]> wrote:
> hello R help
>
> I am trying to analyze a data set that has been collected from a
> hierarchical sampling design. The model should be a mixed model nested
> ANOVA. The purpose of my study is to analyze the variability at each
30 matches
Mail list logo
