R help - Jul 2008 - Correlation structures in gls with repeated measurements

Hi,

I am trying include a spatial correlation structure into a gls model but
keep getting the following error:
> ocean<-read.table("C:\\oceanography2.txt",header=TRUE)
> library(nlme)
> YEAR<-as.factor(ocean$YEAR)
>group<-groupedData(san_dens~temp_grad*pyc_grad*start_thermo|YEAR,data=ocean)
> model1<-gls(san_dens~temp_grad*pyc_grad*start_thermo,data=group)
> model2<-update(model1,corr=corSpher(form=~MIDLONGDEC+MIDLATDEC|YEAR))
Error in getCovariate.corSpatial(object, data = data) :
  Cannot have zero distances in "corSpatial"


The measurements have been repeated over 10 years so the coordinates are
repeated which I assume is the reason for this error. I thought grouping
by YEAR should remove this but it doesnt seem to work, or I am making a
mistake when I am trying to group it? I have noticed a couple of other
posts with the same problem but I have still not managed to solve this.

I would be really grateful if anyone has any ideas or has had this problem
before.


Thanks in advance,
Emma Hazard

Emma,

It looks like there is at least one location that has been sampled at
least twice in some year. A co-worker of me got similar errors. After
checking his data carefully he found a duplicate row. Removing the
duplicate row solved his problem.

HTH,

Thierry


------------------------------------------------------------------------
----
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium 
tel. + 32 54/436 185
Thierry.Onkelinx op inbo.be 
www.inbo.be 

To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to
say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of
data.
~ John Tukey

-----Oorspronkelijk bericht-----
Van: r-help-bounces op r-project.org [mailto:r-help-bounces op r-project.org]
Namens t03eh7 op abdn.ac.uk
Verzonden: woensdag 2 juli 2008 12:29
Aan: r-help op r-project.org
Onderwerp: [R] Correlation structures in gls with repeated measurements

Hi,

I am trying include a spatial correlation structure into a gls model but
keep getting the following error:
> ocean<-read.table("C:\\oceanography2.txt",header=TRUE)
> library(nlme)
> YEAR<-as.factor(ocean$YEAR)
>group<-groupedData(san_dens~temp_grad*pyc_grad*start_thermo|YEAR,data=o
cean)
> model1<-gls(san_dens~temp_grad*pyc_grad*start_thermo,data=group)
> model2<-update(model1,corr=corSpher(form=~MIDLONGDEC+MIDLATDEC|YEAR))
Error in getCovariate.corSpatial(object, data = data) :
  Cannot have zero distances in "corSpatial"


The measurements have been repeated over 10 years so the coordinates are
repeated which I assume is the reason for this error. I thought grouping
by YEAR should remove this but it doesnt seem to work, or I am making a
mistake when I am trying to group it? I have noticed a couple of other
posts with the same problem but I have still not managed to solve this.

I would be really grateful if anyone has any ideas or has had this
problem
before.


Thanks in advance,
Emma Hazard

______________________________________________
R-help op r-project.org mailing list
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.

Thank you, it looks like that was the problem

Emma

> Emma,
>
> It looks like there is at least one location that has been sampled at
> least twice in some year. A co-worker of me got similar errors. After
> checking his data carefully he found a duplicate row. Removing the
> duplicate row solved his problem.
>
> HTH,
>
> Thierry
>
>
> ------------------------------------------------------------------------
> ----
> ir. Thierry Onkelinx
> Instituut voor natuur- en bosonderzoek / Research Institute for Nature
> and Forest
> Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
> methodology and quality assurance
> Gaverstraat 4
> 9500 Geraardsbergen
> Belgium
> tel. + 32 54/436 185
> Thierry.Onkelinx at inbo.be
> www.inbo.be
>
> To call in the statistician after the experiment is done may be no more
> than asking him to perform a post-mortem examination: he may be able to
> say what the experiment died of.
> ~ Sir Ronald Aylmer Fisher
>
> The plural of anecdote is not data.
> ~ Roger Brinner
>
> The combination of some data and an aching desire for an answer does not
> ensure that a reasonable answer can be extracted from a given body of
> data.
> ~ John Tukey
>
> -----Oorspronkelijk bericht-----
> Van: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org]
> Namens t03eh7 at abdn.ac.uk
> Verzonden: woensdag 2 juli 2008 12:29
> Aan: r-help at r-project.org
> Onderwerp: [R] Correlation structures in gls with repeated measurements
>
> Hi,
>
> I am trying include a spatial correlation structure into a gls model but
> keep getting the following error:
>
>> ocean<-read.table("C:\\oceanography2.txt",header=TRUE)
>> library(nlme)
>
>> YEAR<-as.factor(ocean$YEAR)
>
>>group<-groupedData(san_dens~temp_grad*pyc_grad*start_thermo|YEAR,data=o
> cean)
>
>> model1<-gls(san_dens~temp_grad*pyc_grad*start_thermo,data=group)
>
>>
model2<-update(model1,corr=corSpher(form=~MIDLONGDEC+MIDLATDEC|YEAR))
> Error in getCovariate.corSpatial(object, data = data) :
>   Cannot have zero distances in "corSpatial"
>
>
> The measurements have been repeated over 10 years so the coordinates are
> repeated which I assume is the reason for this error. I thought grouping
> by YEAR should remove this but it doesnt seem to work, or I am making a
> mistake when I am trying to group it? I have noticed a couple of other
> posts with the same problem but I have still not managed to solve this.
>
> I would be really grateful if anyone has any ideas or has had this
> problem
> before.
>
>
> Thanks in advance,
> Emma Hazard
>
> ______________________________________________
> R-help at r-project.org mailing list
> 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.
>