similar to: MGCV: Use of irls.reg option

Hi all, I am also confused about the manual: a. The input arguments: wt.method are the weights case weights (giving the relative importance of case, so a weight of 2 means there are two of these) or the inverse of the variances, so a weight of two means this error is half as variable? w (optional) initial down-weighting for each case. init (optional) initial values for the

Dear list, I do apologize if these are basic questions. I am fitting some GAM models using the mgcv package and following the model selection criteria proposed by Wood and Augustin (2002, Ecol. Model. 157, p. 157-177). One criterion to decide if a term should be dropped from a model is if the estimated degrees of freedom (EDF) for the term are close to their lower limit. What would be the

hi Using a ts or tprs basis, I expected gcv to decrease when increasing the basis dimension, as I thought this would minimise gcv over a larger subspace. But gcv increased. Here's an example. thanks for any comments. greg #simulate some data set.seed(0) x1<-runif(500) x2<-rnorm(500) x3<-rpois(500,3) d<-runif(500) linp<--1+x1+0.5*x2+0.3*exp(-2*d)*sin(10*d)*x3

Hello, After reinstalling the whole OS and R as well, I tried to update.packages() and get the follwing error message: concerning the mgcv update: atlas2-base is installed and blas as well (on debian). I haven't found lf77blas, I assume it's a library or something similar associated with blas. any suggestion how to solve that, thanks Martin * Installing *source* package

Hi I just upgraded my debian/stable to R 2.7.1 via apt-get install r-base r-base-core r-base-dev, and then began to update.packages() > update.packages(lib.loc="/usr/local/lib/R/site-library") > update.packages(lib.loc="/usr/lib/R/library") but I get: .... * Installing *source* package 'mgcv' ... ** libs gcc -std=gnu99 -I/usr/share/R/include -fpic -g

Dear useRs, I have used using the excellent mgcv package (version 1.7-12) to create a generalized additive model (gam) including random effects - represented with s(...,bs="re") - on the basis of dialect data. My model contains two random-effect factors (Word and Key - the latter representing a speaker) and I have added both random intercepts and various random slopes for these

I am using gam in mgcv (1.3-22) and trying to use gcv to help with model selection. However, I'm a little confused by the process of assessing GCV scores based on their magnitude (or on relative changes in magnitude). Differences in GCV scores often seem "obvious" with my poisson gams but with negative binomial, the decision seems less clear. My data represent a similar pattern as

Qusetion #1 ********* Model selection in GAM can be done by using: 1. step.gam {gam} : A directional stepwise search 2. gam {mgcv} : Smoothness estimation using GCV or UBRE/AIC criterion Suppose my model starts with a additive model (linear part + spline part). Using gam() {mgcv} i got estimated degrees of freedom(edf) for the smoothing splines. Now I want to use the functional form of my model

Greetings I run into a compilation error when updating to mgcv_1.1-7 in R 2.0.0 on OS X 10.3. Note that other pacakges have compiled nicely. Some details are given below, but in short it looks like it's seeking for /usr/local/lib/powerpc-apple-darwin6.8/3.4.2/ which I don't have. But I do have /usr/lib/gcc/darwin/3.3 i.e a lower version of GCC in a different directory. More

Christoph Lehman had problems with seperated data in two-class logistic regression. One useful little trick is to penalize the logistic regression using a quadratic penalty on the coefficients. I am sure there are functions in the R contributed libraries to do this; otherwise it is easy to achieve via IRLS using ridge regressions. Then even though the data are separated, the penalized

Dear List, I'm using gam in a multiple imputation framework -- specifying the knot locations, and saving the results of multiple models, each of which is fit with slightly different data (because some of it is predicted when missing). In MI, coefficients from multiple models are averaged, as are variance-covariance matrices. VCV's get an additional correction to account for how

Hi Paul, Take a look at gam() from package mgcv (gam = generalized additive models), maybe this will help you. GAMs can work with other distributions as well. Generalized additive models consist of a random component, an additive component, and a link function relating these two components. The response Y, the random component, is assumed to have a density in the exponential family. I am not sure

For summary(GAM) in the mgcv package smooth the degrees of freedom for the F value for test of smooth terms are the rank of covariance matrix of \hat{beta} and the residuals df. I've noticed that in a lot of GAMs I've fit the rank of the covariance turns out to be 9. In Simon Wood's book, the rank of covariance matrix is usually either 9 or 99 (pages 239-230 and 259). Can anyone

Hi there, I was wondering if by the way the isotropic bivariate function works in the mgcv package, one can use highly correlated coordinates (given the shape of the study area) without worrying about the potential problems of correlation between explanatory variables, i.e., does s(LON, LAT) deal with that by considering their combined effect? Although this sounds more like a statistical

Updating my 1.8.0 R installation (>update.packages() ) I obtain the following (SORRY FOR THE LENGTH OF THE LOG BUT IT HELPS!!!): ................ downloaded 135Kb KernSmooth : Version 2.22-11 in /usr/lib/R/library Version 2.22-12 on CRAN Update (y/N)? y mgcv : Version 0.9-3.1 in /usr/lib/R/library Version 0.9-6 on CRAN Update (y/N)? y trying URL

Greetings, I am learning R My machine has these; CPU: 3cores amd64 OS pure-64bit CBLFS liux compiled from sources (kernel 3.2.1, gcc-4.6.2 R-2.15 When I compiled R the compiler spewed out lines like these:- make[3]: Entering directory `/tmp/RtmpiHdDJy/R.INSTALL472339eeb23a/mgcv/src' gcc -m64 -std=gnu99 -I/home/Rman/R-2.15.0/include -DNDEBUG - I/usr/local/atlas/include

Hi, I have been trying to fit an un-penalised gam in mgcv (in order to get more reliable p-values for hypothesis testing), but I am struggling to get the model to fit sucessfully when I add in a te() interaction. The model I am trying to fit is: gam(count~ s(x1, bs = "ts", k = 4, fx = TRUE) + s(x2, bs = "ts", k = 4, fx = TRUE) + te(x2, x3, bs =

Hi all R users ! I'm using gamm function from Simon Wood's mgcv package, to fit a spatial regression Generalized Additive Mixed Model, as covariates I have the geographical longitude and latitude locations of indexed data. I include a random effect for each district (dist) so the code is fit <- gamm(y~s(lon,lat,bs="tp", m=2)+offset(log(exp.)), random=list(dist=~1),

I would like to fit a logistic regression using a smothing spline, where the spline is a piecewise cubic polynomial. Is the knots option used to define the subintervals for each piece of the cubic spline? If yes and there are k knots, then why does the coefficients field in the returned object from gam only list k coefficients? Shouldn't there be 4k -4 coefficients? Sincerely, Bill

Hi, I am trying to build GAM with linear constraints, for a general link function, not only identity. If I understand it correctly, the function pcls() can solve the problem, if the smoothness penalties are given. What I need is to incorporate the constraints before calculating the penalties. Can this be done in R? Any help would be greately appreciated. -- View this message in context: