[R] extending family() objects
Dear R-help, The family collection of objects is very useful since I can perform some of the calculations involved when fitting glms easily in a vectorized manner. I would like to extend them in the following manner: I want to supply a vector of family names such as c(poisson, gamma,beta), and an indexing vector such as c(1,1,1,1,1,1,2,2,2,2,3,3,3,3) ( integer elements that go up to the length of the vector of family names). I then want to construct a new type of family object, new.family, such that when calling any of its functions, eg. family.new$variance(mu), it will apply a different function according to the indexing vector. I could do this quick-and-dirty using loops but I would really like not to loose much of the speed obtained through vectorised calculations. Any ideas? Thanks Simon Bond. - /\ \ /ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL X - AGAINST MS ATTACHMENTS / \ http://www.gnu.org/philosophy/no-word-attachments.html __ 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] negative binomial in GEE
Dear R-help, I was recently wanting to use GEE with the negative binomial family. It seems that this is lacking in the otherwise excellent implementations of the GEE methodology ( packages: gee, yags, geepack). I would have thought it a simple step to allow the creation of a family, i.e providing the link function (log mu) and the variance function (mu + mu^2/theta) , assuming theta is specified; and then to let the gee code do the business in a similar fashion to glm. However the gee codes all seem to rely a set of C routines with either the normal, poisson, or binomial, hardwired in. Does anyone have any suggestions for a way round this problem for the future (I had to resort to using Stata), or maybe more realistically, how much work it would take to build an extendible version of the gee algorithm? thanks Simon Bond. __ 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] font sizes
Having experimented with both a sun workstation and a PC, changing pointsize within the PC does have the desired effect, but it does nothing within the Sun. Unforutnately, the PC won't let me load the workspace I want (which I normally access through the sun) due to `lazy loading' errors. Thinking more generally, although I think the ability to fine tune R graphics is excellent, even superlative, what I would find really useful is means to load a whole load of graphical settings in one go; one setting to look at on-screen, one setting for written reports, one setting for slides. Can anyone suggest a good way of going about this. thanks Simon On Tue, 15 Mar 2005, Uwe Ligges wrote: Simon.Bond wrote: I'm trying to use the pdf() function, and would like to increase the font size for slide-presentation purposes. Changing the argument `pointsize' doesn't seem to do anything. Anyone come across this or know what to do? It does, e.g. compare pointsize=8 / pointsize=14 If you want something different, maybe setting argument cex (and friends) in par() does what you want. See ?par. Uwe Ligges thanks Simon Bond. - /\ \ /ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL X - AGAINST MS ATTACHMENTS / \ http://www.gnu.org/philosophy/no-word-attachments.html __ 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 - /\ \ /ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL X - AGAINST MS ATTACHMENTS / \ http://www.gnu.org/philosophy/no-word-attachments.html __ 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] impenetrable warning
Dear R-help, Can anyone explain the meaning of the warning, Singular precision matrix in level -1, block 1 ? Or how to track down where it comes from? More precisely, using the nlme package, I'm issued with the warning itt2 - lme(lrna~rx.nrti+lbrna, random=~1|patid, cor=corExp(form=~days|patid,nugget=T), weights=varPower( form=~lbrna),data=rna3) Warning messages: 1: Singular precision matrix in level -1, block 1 2: Singular precision matrix in level -1, block 1 the output is: Linear mixed-effects model fit by REML Data: rna3 Log-restricted-likelihood: -4990.142 Fixed: lrna ~ rx.nrti + lbrna (Intercept) rx.nrtiZDV+3TC rx.nrtiZDV+ABC lbrna 2.6597552 0.8589514 0.4259504 0.2929222 Random effects: Formula: ~1 | patid (Intercept) Residual StdDev:1.251113 0.2105329 Correlation Structure: Exponential spatial correlation Formula: ~days | patid Parameter estimate(s): range nugget 204.6422150 0.3349336 Variance function: Structure: Power of variance covariate Formula: ~lbrna Parameter estimates: power 0.9733358 Number of Observations: 2390 Number of Groups: 124 Thanks, Simon Bond. __ [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
[R] Interpretation of poly(x, y, degree=2) coefficients
Does anyone know of a function that turns the coefficients of lm( y~ poly(x1 , x2, degree=2)) into something that can be interpreted easily. I was think along the lines of the matrix representation of quadratic forms: (x-mu)'A(x-mu) +k , and finding the eigenvectors/values of A, and the vector mu, but anything that allows me to visualise a contour plot would be great. Thanks Simon Bond __ [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] update/offset
I've noticed that the update() function does not seem to work correctly when offset(..) terms are there: update(modelwithoffset, .~.-afixedeffect) drops the offset term. I'm using this with a negbin model, but I think it goes wrong with the update.formula() function. update.formula function (old, new, ...) { env - environment(as.formula(old)) tmp - .Internal(update.formula(as.formula(old), as.formula(new))) out - formula(terms.formula(tmp, simplify = TRUE)) environment(out) - env return(out) } environment: namespace:base The .Internal command in the code is where my efforts grind to a halt. Any suggestions (apart from redefine the new model from scratch) ? Thanks Simon Bond. __ [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] summary.nlme
Hi, I'm trying to work out how the nlme function estimates the variances of the fixed effects parameters, so I tried to look at the code for these functions: summary.nlme, summary.lme, MEestimate. MEestimate Error: Object MEestimate not found summary.nlme Error: Object summary.nlme not found summary.lme Error: Object summary.lme not found So R says they don't exist, although summary(AnNlmeObject) works fine. What's going on? Simon Bond. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help