[R] removing dropouts from survival analysis

Dear All. Apologies for posting a question regarding survival analysis, and not R, to the R-help list. In the past I received the best advices from the R community. The random censorship model (the censoring times independent of the failure times and vice versa) is one of the fundamental

[R] independent censoring

Dear All, Independent censoring is one of the fundamental assumptions in the survival analysis. However, I cannot find any test for it or any paper which discusses how real that assumption is. I would be grateful if anybody could point me to some useful references. I have found the

Re: [R] survival survfit with newdata

To: dkrsta...@hotmail.com Subject: Re: [R] survival survfit with newdata Date: Thu, 17 May 2012 00:52:55 -0400 On May 16, 2012, at 5:08 PM, Damjan Krstajic wrote: Dear all, I am confused with the behaviour of survfit with newdata option. Yes. It has the same behavior as any other

[R] survival survfit with newdata

Dear all, I am confused with the behaviour of survfit with newdata option. I am using the latest version R-2-15-0. In the simple example below I am building a coxph model on 90 patients and trying to predict 10 patients. Unfortunately the survival curve at the end is for 90 patients. Could

[R] (bug?) in survfit in R-2-15-0

Dear all,   I am using glmnet + survival and due to the latest release of glmnet 1.7.4 I was forced to use the latest version of R 2.15.0. My previous version of R was 2.10.1. I changed glmnet version and R version and when I started to get weird results I was not sure where the bug was.  

[R] estimating survival times with glmnet and coxph

Dear all, I am using glmnet (Coxnet) for building a Cox Model and to make actual prediction, i.e. to estimate the survival function S(t,Xn) for a new subject Xn. If I am not mistaken, glmnet (coxnet) returns beta, beta*X and exp(beta*X), which on its own cannot generate S(t,Xn). We miss baseline

Re: [R] differences between 1.7 and 1.7.1 glmnet versions

Thank you very much. I will use 1.7.1 version. Have you had time to look at the issue regarding a weird behaviour of multinomial glmnet when alpha=0? I posted it to r-help more than two weeks ago and maybe you missed it. Damjan Krstajic Subject: Re: differences between 1.7 and 1.7.1 glmnet

[R] differences between 1.7 and 1.7.1 glmnet versions

Dear All,   I have found differences between glmnet versions 1.7 and 1.7.1 which, in my opinion, are not cosmetic and do not appear in the ChangeLog. If I am not mistaken, glmnet appears to return different number of selected input variables, i.e. nonzeroCoef(fit$beta[[1]]) differes between

[R] bug in glmnet 1.7.1 for multinomal when alpha=0?

Dear all, If I am not mistaken, I think that I have found a bug in glmnet 1.7.1 (latest version) for multinomial when alpha=0. Here is the code library(glmnet) Loading required package: Matrix Loading required package: lattice Loaded glmnet 1.7.1 x=matrix(rnorm(40*500),40,500)

Re: [R] survival: ridge log-likelihood workaround

to be part of summary for ridge coxph and I look forward to be corrected. On my behalf I am prepared in my spare time to write the code so that the summary for ridge coxph does not return NULL and that the statistics printed in summary for ridge coxph are based on published papers. Damjan

[R] survival: ridge log-likelihood workaround

Dear all, I need to calculate likelihood ratio test for ridge regression. In February I have reported a bug where coxph returns unpenalized log-likelihood for final beta estimates for ridge coxph regression. In high-dimensional settings ridge regression models usually fail for lower values of

[R] survival - summary and score test for ridge coxph()

It seems to me that summary for ridge coxph() prints summary but returns NULL. It is not a big issue because one can calculate statistics directly from a coxph.object. However, for some reason the score test is not calculated for ridge coxph(), i.e score nor rscore components are not included

[R] confidence intervals for non-linear regression

() is kind of black box. Any references to the literature or R packages would be very welcome. Thanks in advance. DK -- Damjan Krstajic Director Research Centre for Cheminformatics Belgrade, Serbia

[R] scientific (statistical) foundation for Y-RANDOMIZATION in regression analysis

-- Damjan Krstajic Director Research Centre for Cheminformatics Belgrade, Serbia -- _ Tell us your greatest, weirdest

[R] survival - ratio likelihood for ridge coxph()

It seems to me that R returns the unpenalized log-likelihood for the ratio likelihood test when ridge regression Cox proportional model is implemented. Is this as expected? In the example below, if I am not mistaken, fit$loglik[2] is unpenalized log-likelihood for the final estimates of

[R] Introducing R to statisticians

Dear all, I will present R language and R software environment to the Statistical Society of Serbia. As I will doing it to professional statisticians it seems unneccesary to me to present them how R language works in details. I am more interested to present them with the latest facts regarding

[R] statistics and R package for election results

Dear all, Is there any R package which would help in analysing election results between two elections? Does anybody know any good papers which are related to this field? I am a statistician and my main research area so far has been regression and classification modelling. The analysis of two

[R] R survival package error message - bug?!

Dear all, I have encountered a weird behaviour in R survival package which seems to me to be a bug. The weird behaviour happens when I am using 100 variables in the ridge function when calling coxph with following formula Surv(time = futime, event = fustat, type = right) ~ ridge(X1, X2,

[R] coxph (survival) paper from '82

of Statistics 10, 1100-1120 I would be really grateful if someone could help me with the electronic copy of the paper. Writing to the R-help emailing list was my last resort. Thanks in advance. With kind regards DK Damjan Krstajic Director Research Centre for Cheminformatics