similar to: Re: frailty models in survreg() -- survival package (PR#2934)

SEE ALSO ORIGINAL POSTING IN PR#2933 On May 6, 2003 03:58 pm, Thomas Lumley wrote: > > Looking at a wider context in the code > > pfun <- function(coef, theta, ndeath) { > if (theta == 0) > list(recenter = 0, penalty = 0, flag = TRUE) > else { > recenter <- log(mean(exp(coef))) > coef <- coef - recenter

I am confused on how the log-likelihood is calculated in a parametric survival problem with frailty. I see a contradiction in the frailty() help file vs. the source code of frailty.gamma(), frailty.gaussian() and frailty.t(). The function frailty.gaussian() appears to calculate the penalty as the negative log-density of independent Gaussian variables, as one would expect: >

Dear All, I am attempting to make predictions based on a survreg() model with some censoring and a frailty term, as below: predict works fine on the original data, but not if I specify newdata. # a model with groups as fixed effect model1 <- survreg(Surv(y,cens)~ x1 + x2 + groups, dist = "gaussian") # and with groups as a random effect fr <- frailty(groups,

The combination of survreg + gamma frailty = invalid model, i.e., the example that you quote. I did not realize that this had been added to the survreg help file until very recently. I will try to fix the oversight. Other, more detailed documentation states that Gaussian frailty + AIC is the only valid random effects choice for survreg. Details: frailty(x) with no optional

I have the following questions about the variance of the random effects in the survreg() function in the survival package: 1) How can I extract the variance of the random effects after fitting a model? For example: set.seed(1007) x <- runif(100) m <- rnorm(10, mean = 1, sd =2) mu <- rep(m, rep(10,10)) test1 <- data.frame(Time = qsurvreg(x, mean = mu, scale= 0.5, distribution =

Dear R-users, I have noticed small discrepencies in the reported estimate of the variance of the frailty by the print method for survreg() and the 'theta' component included in the object fit: # Examples in R-2.6.2 for Windows library(survival) # version 2.34-1 (2008-03-31) # discrepancy fit1 <- survreg(Surv(time, status) ~ rx + frailty(litter), rats) fit1 fit1$history[[1]]$theta

A couple of clarifications for you. 1. I write mixed effects Cox models as exp(X beta + Z b), beta = fixed effects coefficients and b = random effects coefficients. I'm using notation that is common in linear mixed effects models (on purpose). About 2/3 of the papers use exp(X beta)* c, i.e., pull the random effects out of the exponent. Does it make a difference? Not much: b will be

Hi, I would like to fit parametric survival models to time-to-event data that are left truncated. I have checked the help page for survreg and looked in the R-help archive, and it appears that the R function survreg from the survival library (version 2.16) should allow me to take account of left truncation. However, when I try the command

Does someone know the rules by which 'coxph' returns 'frail', the predicted frailty terms? In my test function: ----------------------------------------------- fr <- function(){ #testing(frailty terms in 'survival' require(survival) dat <- data.frame(exit = 1:6, event = rep(1, 6), x = rep(c(0, 1), 3),

Hello, I was hoping that someone could answer a few questions for me (the background is given below): 1) Can the coxph accept an interaction between a covariate and a frailty term 2) If so, is it possible to a) test the model in which the covariate and the frailty appear as main terms using the penalized likelihood (for gaussian/t frailties) b)augment model 1) by stratifying on the variable that

Dear R-Helpers, I am hoping to perform survival analyses using the "ex-Gaussian" distribution. I understand that the ex-Gaussian is a convolution of exponential and Gaussian distributions for survival data. I checked the "survreg.distributions" help and saw that it is possible to mix pre-defined distributions. Am I correct to think that the following code makes the

Dear all, I would have some questions on the coxph function for survival analysis, which I use with frailty terms. My model is: mdcox<-coxph(Surv(time,censor)~ gender + age + frailty(area, dist='gauss'), data) I have a very large proportion of censored observations. - If I understand correctly, the function mdcox$frail will return the random effect estimated for each group on the

Dear All, The "coxph" function in the "survival" package allows multiple frailty terms. In all the examples I saw, however, the frailty terms are nested. What will happen if I have non-nested (that is, cross-classified) frailties in the model? Will the model still work? Do I need to take special cares when specifying these models? Thanks! Shige [[alternative HTML

Hello, I'm trying to estimate a cox model with a frailty variable and time-dependent covariate (below there is the statement I use and the error message). It's seems to be impossible, because every time I add the time-dependent covariate the model doesn't converge. Instead, if I estimate the same model without the time-dependent covariate it's converge. I'd like knowing if

> I was wondering why survreg (in survival package) can not handle > three-way interactions. I have an AFT ..... You have given us no data to diagnose your problem. What do you mean by "cannot handle" -- does the package print a message "no 3 way interactions", gives wrong answers, your laptop catches on fire when you run it, ....? Also, make sure you read

Hi everybody. I'm trying to fit a weibull survival model with a spline basis for the predictor, using the survival library. I've noticed that it doesn't seem to be possible to use the aic method to choose the degrees of freedom for the spline basis in a parametric regression (although it's fine with the cox model, or if the degrees of freedom are specified directly by the user),

Dear R users: I'm generating the following survival data: set.seed(123) n=200 #sample size x=rbinom(n,size=1,prob=.5) #binomial treatment v=rgamma(n,shape=1,scale=1) #gamma frailty w=rweibull(n,shape=1,scale=1) #Weibull deviates b=-log(2) #treatment's slope t=exp( -x*b -log(v) + log(w) ) #failure times c=rep(1,n) #uncensored indicator id=seq(1:n) #individual frailty indicator

Hi, I have a question about the output for variance of random effect from a gamma frailty model using coxph in R. Is it the vairance of frailties themselves or variance of log frailties? Thanks. Guanghui

Dear list, I want o fit a shared gamma frailty model with the frailty specification in the survival package. I have partly left-truncated data and time-dependent covariates. Is it possible to combine these two things in the frailty function. Or are the results wrong if I use data in the start-stop-formulation which account for delayed entry? Is the frailty distribution updated in the

It can't be done with the current code. In a nutshell, you are trying to use a feature that I never got around to coding. It's been on my "to do" list, but may never make it to the top. Terry