Re: [R] LME correlation structures: user defined
Hi, I noticed your old posts whilst tidying up my mailbox. I am struggling trying to understand the code to define new structures - part of the real work is hidden away in C code and I don't speak C very well. Have either of you successfully implemented any alternative structures? (At one stage I ws interested in trying Toeplitz forms, and now have a need for compound symmetry forms where the rho depends on a factor.) Have you anything beyond the source that might help? - for example a more richly commented example, or notes on what the C (and R) functions actually do. Regards, Steve. Date sent: Wed, 23 Mar 2005 10:56:46 +1100 To: "Michael Jerosch-Herold" <[EMAIL PROTECTED]>, R-help@stat.math.ethz.ch From: Simon Blomberg <[EMAIL PROTECTED]> Subject:Re: [R] LME correlation structures: user defined Copies to: > >Let me modify my question about user-defined covariance structures > >for LME models: Can somebody tell me how I can see the code for the > >definition of the correlation structures that come with the NLME > >package. Specifically I like to see the code for the functions coef, > >corMatrix, and intialize for any of the pre-defined correlation > >structures, and use this as a template to define a new correlation > >structure. So how do I see e.g. the code for the method initialize > >for the correlation structure corExp or corARMA? > > I'm interested in this too. Go to CRAN and download the source code > for the nlme package. ungzip and untar it. Got to the R subdirectory. > Inside that is a file called corStruct.R with all the method > definitions for the built-in corStruct classes. Reading those should > help. > > Cheers, > > Simon. > > > > > >thank you in advance! > > > >Michael Jerosch-Herold > > > >PS: Oh, and if somebody could still send me example code for a user > >defined correlation structure I would much appreciate it, as my > >previous requests for help have not yielded any response. > > > >__ > >R-help@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 > > > -- > Simon Blomberg, B.Sc.(Hons.), Ph.D, M.App.Stat. > Visiting Fellow > School of Botany & Zoology > The Australian National University > Canberra ACT 0200 > Australia > > T: +61 2 6125 8057 email: [EMAIL PROTECTED] > F: +61 2 6125 5573 > > CRICOS Provider # 00120C > > __ > R-help@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 Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192/5764 / 0161 276 5785 __ R-help@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
[R] lme/gls heteroscedastic correlation structure
Greetings, We want to fit lme/gls models with a compound symmetry variance structure, but where the ICC takes different values across different groups of clusters. (Specifically a two arm clustrered trial with different ICC in each arm) Equivalent models can be fitted with heteroscedastic variance functions, but these have positivity constraints which we want to avoid and we would prefer to use the correlation structure directly. Is there a way to do this? Writing a new variance structure seems rather non-trivial - but maybe I haven't tried hard enough.. Any ideas? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192/5764 / 0161 276 5785 __ R-help@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
Re: [R] R2.1.0: Bug in list.files
Yes I missed the NEWS entry - or rather didn't realise its significance.
So the "bug" was in the previous version and my old code which worked
but shouldn't have.
Thanks for the replies - rapid and to the point as usual.
Steve.
Date sent: Thu, 12 May 2005 10:45:03 +0200
From: Uwe Ligges <[EMAIL PROTECTED]>
Organization: Fachbereich Statistik, Universitaet Dortmund
To: [EMAIL PROTECTED]
Copies to: R-help@stat.math.ethz.ch
Subject:Re: [R] R2.1.0: Bug in list.files
> Steve Roberts wrote:
>
> > R2.0.1 (MS Windows)
> >
> >
> >>list.files(myloc,"*.zip",full=T)
> >
> > [1] "P:/SARsoftware/Rlibraries/gnlm_0.1.zip"
> > [2] "P:/SARsoftware/Rlibraries/lms2_0.2.zip"
> >
> >
> > R2.1.0:
> >
> >
> >>list.files(myloc,"*.zip",full=T)
> >
> > Error in list.files(path, pattern, all.files, full.names, recursive)
> > :
> > invalid 'pattern' regular expression
> >
> > Bug? or have I missed something
>
> You missed to read the NEWS that tells you:
>
> o The regular expression code is now based on that in glibc
> 2.3.3. It has stricter conformance to POSIX, so metachars such as
> { } + * may need to be escaped where before they did not (but
> could have been).
>
>
> Probably you want
>
> list.files(pattern = "\\.zip$", full.names = TRUE)
>
> Uwe Ligges
>
>
> > Steve.
> > Dr Steve Roberts
> > [EMAIL PROTECTED]
> >
> > Senior Lecturer in Medical Statistics,
> > CMMCH NHS Trust and University of Manchester Biostatistics Group,
> > 0161 275 5192/5764 / 0161 276 5785
> >
> > __
> > R-help@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
>
Dr Steve Roberts
[EMAIL PROTECTED]
Senior Lecturer in Medical Statistics,
CMMCH NHS Trust and University of Manchester Biostatistics Group,
0161 275 5192/5764 / 0161 276 5785
__
R-help@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
[R] R2.1.0: Bug in list.files
R2.0.1 (MS Windows) > list.files(myloc,"*.zip",full=T) [1] "P:/SARsoftware/Rlibraries/gnlm_0.1.zip" [2] "P:/SARsoftware/Rlibraries/lms2_0.2.zip" R2.1.0: > list.files(myloc,"*.zip",full=T) Error in list.files(path, pattern, all.files, full.names, recursive) : invalid 'pattern' regular expression Bug? or have I missed something Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192/5764 / 0161 276 5785 __ R-help@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
[R] lme RE variance computation
As I understand it lme (in R v1.9.x) estimates random effect variances on a log scale, constraining them to be positive. Whilst this seems sensible, it does lead to apparently biased estimates if the variance is actually zero - which makes our simulation results look strange. Whilst we need to think a bit deeper about it - I still haven't got my head around what a negative variance could mean - does anyone know a way to take away the contraint and allowing zero or negative variances? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Help with plotmath
No that isn't it I'm afraid - my expression is pasted together from various bits and pieces and is at some stage a string variable. I needed parse as Brian pointed out Thanks anyway, Steve. From: "Petr Pikal" <[EMAIL PROTECTED]> To: "Steve Roberts" <[EMAIL PROTECTED]> Date sent: Thu, 10 Jun 2004 13:44:21 +0200 Subject:Re: [R] Help with plotmath Copies to: [EMAIL PROTECTED] Priority: normal > Hi > > On 10 Jun 2004 at 12:28, Steve Roberts wrote: > > > There must be a simple answer. I want to plot an expression, > > where the expression is held in a string variable. The "obvious" > > solution along the lines of > > > > ex<-"x^2" > > plot( c(0,1), c(0,1), main=as.expression(ex) ) > > Plotmath example led me to > > ex<-expression(x^2) > plot( c(0,1), c(0,1), main=(ex) ) > > hope it is what you want. > > Cheers > Petr > > > > > > gives me the a title x^2 - ie doesn't treat it like an expression. I > > suspect I don't understand expressions properly. For the real problem > > I do need to have my expression string in a variable as it is > > data-dependent. > > > > Can someone tell me the magic words? > > > > Steve > > Dr Steve Roberts > > [EMAIL PROTECTED] > > > > Senior Lecturer in Medical Statistics, > > CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 > > 275 5192 / 0161 276 5785 > > > > __________ > > [EMAIL PROTECTED] mailing list > > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! > > http://www.R-project.org/posting-guide.html > > Petr Pikal > [EMAIL PROTECTED] > > Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Help with plotmath
There must be a simple answer. I want to plot an expression, where the expression is held in a string variable. The "obvious" solution along the lines of ex<-"x^2" plot( c(0,1), c(0,1), main=as.expression(ex) ) gives me the a title x^2 - ie doesn't treat it like an expression. I suspect I don't understand expressions properly. For the real problem I do need to have my expression string in a variable as it is data-dependent. Can someone tell me the magic words? Steve Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] 2 lme questions
Greetings, 1) Is there a nice way of extracting the variance estimates from an lme fit? They don't seem to be part of the lme object. 2) In a series of simulations, I am finding that with ML fitting one of my random effect variances is sometimes being estimated as essentially zero with massive CI instead of the finite value it should have, whilst using REML I get the expected value. I guess it is a numerical/optimisation problem but don't know enough about the lme fitting algorithm to know which tollerance/scale parameter to mess about with. Any suggestions where to start? Thanks, Steve. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] DLLs and the Floating Point Control Word.
Greetings. One for the developers I guess... I am having problems in using a (non-recommended) Fortran compiler (Salford ftn95 Windoze), and the crashes do seem to be associated with the ftn95-dervived DLL changing the Floating Point Control Word. The compiler people are suggesting (and I paraphrase!) that if R minds what the FPCW is it is up to R to make sure it is the value it wants and R should check/reset on returning from the DLL call. There seems to me to be some logic in this position - if only to make life easier for those of us who don't really know what a FPCW is! Having to add unintelligable code to reset the FPCW at the end of each fortran routine is inelegant. It probably isn't as easy as it sounds - I guess there are multi-platform issues? Is this a sensible suggestion to add to the wish list? Keep up the good work Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] RGui (Windows) crashes after use of a Salford Fortran DLL
Anybody out there successfully using the Salford Fortran compilers with R? I have created a DLL using the Salford FTN95 compiler and it works in as far I can dyn.load it, run the routines and get the right answers back. Unfortunately subsequently, sometime later, the Rgui crashes (access violation I think from the DrWatson log). The crashes depend on whether or not I paste the code as one big chunk, or as little chunks, and sometimes I can do a few things before it crashes and other times not - but if I do exactly the same things the crashes are reproducible. The crashes are sometimes in the DLL and sometimes in pure R code. R1.7.1 and 1.8.1 seem to behave identically. Windows 2000 Pro. I do get the dyn.load warning about the DLL changing the FPU control word. Does this ring any bells with anyone? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Bug in cor.test - Spearman
Greetings. There seems to be a problem with the P-value computation in the cor.test with method="spearman". In R1.8.0 (MS Windows) I seem to be getting intermittently nonsense P-values, but the rho's are OK. I can get this reproducibly with the toy example attached where the first use is OK and subsequent calls with the same data give nonsense. (I have also seen the problem without the warning about duplicated values.) The toy example behaves correctly under 1.7.1. Steve. ---Script--- x<-1:100 y<-rep(c(2,3,4,5),25) cor.test(x,y,m="p") cor.test(x,y,m="s") cor.test(x,y,m="s") cor.test(x,y,m="k") ---Output-- R : Copyright 2003, The R Development Core Team Version 1.8.0 (2003-10-08) R is free software and comes with ABSOLUTELY NO WARRANTY. You are welcome to redistribute it under certain conditions. Type 'license()' or 'licence()' for distribution details. R is a collaborative project with many contributors. Type 'contributors()' for more information. Type 'demo()' for some demos, 'help()' for on-line help, or 'help.start()' for a HTML browser interface to help. Type 'q()' to quit R. > x<-1:100 > y<-rep(c(2,3,4,5),25) > cor.test(x,y,m="p") Pearson's product-moment correlation data: x and y t = 0.3837, df = 98, p-value = 0.702 alternative hypothesis: true correlation is not equal to 0 95 percent confidence interval: -0.1588952 0.2333745 sample estimates: cor 0.03873177 > cor.test(x,y,m="s") Spearman's rank correlation rho data: x and y S = 160195, p-value = 0.7015 alternative hypothesis: true rho is not equal to 0 sample estimates: rho 0.03873177 Warning message: p-values may be incorrect due to ties in: cor.test.default(x, y, m = "s") > cor.test(x,y,m="s") Spearman's rank correlation rho data: x and y S = 160195, p-value = < 2.2e-16 alternative hypothesis: true rho is not equal to 0 sample estimates: rho 0.03873177 Warning message: p-values may be incorrect due to ties in: cor.test.default(x, y, m = "s") > cor.test(x,y,m="k") Kendall's rank correlation tau data: x and y z = 0.5132, p-value = 0.6078 alternative hypothesis: true tau is not equal to 0 sample estimates: tau 0.03481553 > Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] What a wonderful list (was Strange and disturbing bug)
What a wonderful set of people you are! - I solved by own problem, deleted the message and got 3 replies before I realised I'd hit "send" rather then "delete". Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Strange and disturbing bug
How about this one? If I set a variable in a data.frame with a two- part name including a dot (say y.pair), and if the variable with the name of the first part (y) doesn't but I ask for it's value I get the value of the two-part name. Ie set fred$x.pair and print the value of fred$x it gives me the value I set to fred$x.pair. Weird and somewhat disturbing! Maybe an example makes it clearer: Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Profile on optim/nlm
Greetings, Before I reinvent the wheel has anyone done a profile function for a fit using optim (or nlm)? (like the buggy profile.ms in S+) . It seems a bit tricky as the function to be minimised has to have arguments corresponding to the variables being fitted - which is one less than the function provided to optim()... I guess you can create another function on the fly somehow. Cheers, Steve. Dr Steve Roberts [EMAIL PROTECTED] Biostatistics Group, School of Epidemiology and Health Sciences, 2nd Floor,Stopford Building, University of Manchester. M13 9PT. Tel: 0161 275 5192 FAX: 0161 275 5205 __ [EMAIL PROTECTED] mailing list http://www.stat.math.ethz.ch/mailman/listinfo/r-help
