search for: londonmet

...le binding for global variable 'own.linkinv' make.link.gamlss : mu.eta: no visible binding for global variable 'own.mu.eta' make.link.gamlss : valideta: no visible binding for global variable 'own.valideta' Thanks Mikis Stasinopoulos Companies Act 2006 : http://www.londonmet.ac.uk/companyinfo

Hi, I'm a bit puzzled by the gamlss fitting of exponential and lognormal data. Gamlss seems to think that exponentially distributed data fits better with a lognormal distribution, and vice versa. For example, X <- rexp(1000) Gexp <- gamlss(X~1,family=EXP) # X~1 is X tilde 1 GAMLSS-RS iteration 1: Global Deviance = 2037.825 GAMLSS-RS iteration 2: Global Deviance = 2037.825 Glno

Dear all and Mikis I have the opportunity to compare fits with the 'classical' glm and gamlss and no smoother of any kind just the same model formula (both with the binomial family). I get exactly the same coefficients but very different residuals, gamlss giving residuals which are extremely close to 'normal' and glm very far... How can this be ? Thanks in advance for

...0 From: Dimitrios Stasinopoulos <dmh.stasinopoulos at gmail.com> To: r-help at r-project.org Cc: Nik ntuzov.com> Subject: [R] . Package gamlss used inside foreach() and %dopar% fails to find an object (Nik Tuzov) Message-ID: <424DC535-4CAF-4DD5-8BD5-EB19BF6F7247 at staff.londonmet.ac.uk> Content-Type: text/plain; charset="utf-8" Dear Nik Try the following code loo_predict.mu <- function(model.obj, input.data) { yhat <- foreach(i = 1 : nrow(input.data), .packages="gamlss", .combine = rbind) %dopar% { updated.model.obj <- update(mode...

...= input.data[i,], type = "response") } return(data.frame(result = yhat[, 1], row.names = NULL)) } par.run <- loo_predict.mu(model3, input.processed.cut) The predict command in this case also need the old data. Thanks Mikis Prof Dimitrios Mikis Stasinopoulos stasinom at staff.londonmet.ac.uk [[alternative HTML version deleted]]

Dear all I am trying to create a package in R 1.9.0 and I a getting an error message which I do not understand. (I am using R in Windows XP and 2000) For example the following works well in 1.8.1 C:\Program Files\R\rw1081\src\gnuwin32>make pkg-gamlss ---------- Making package gamlss ------------ adding build stamp to DESCRIPTION installing inst files installing indices not zipping

.../r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > End of R-help Digest, Vol 128, Issue 29 > *************************************** > Companies Act 2006 : http://www.londonmet.ac.uk/companyinfo [[alternative HTML version deleted]]

Greetings, My question is more algorithmic than prectical. What I am trying to determine is, are the GAM algorithms used in the mgcv package affected by nonnormally-distributed residuals? As I understand the theory of linear models the Gauss-Markov theorem guarantees that least-squares regression is optimal over all unbiased estimators iff the data meet the conditions linearity,

I have a two part question one about statistical theory and the other about implementations in R. Thank you for all help in advance. (1) Am I correct in understanding that Heteroscedasticity is a problem for Generalized Additive Models as it is for standard linear models? I am asking particularly about the GAMs as implemented in the mgcv package. Based upon my online search it seems that some

Dear all I am trying to build a package in Windows. I use the following command (which it used to work with previous versions ) and I am getting the following error #-------------------------------------------------------------------------------------------------------------- C:\PROGRA~1\R\rw2000\bin>Rcmd build --binary --use-zip C:\PROGRA~1\R\rw2000\src\library\gamlss * checking for