Re: [Rd] Bootstrapping stepAIC() with glm.nb()

2007-02-23 Thread Prof Brian Ripley
On Fri, 23 Feb 2007, Dimitris Rizopoulos wrote: > > - Original Message - From: "Prof Brian Ripley" > <[EMAIL PROTECTED]> > To: "Dimitris Rizopoulos" <[EMAIL PROTECTED]> > Cc: > Sent: Friday, February 23, 2007 1:40 PM > Subject: Re:

Re: [Rd] Bootstrapping stepAIC() with glm.nb()

2007-02-23 Thread Dimitris Rizopoulos
- Original Message - From: "Prof Brian Ripley" <[EMAIL PROTECTED]> To: "Dimitris Rizopoulos" <[EMAIL PROTECTED]> Cc: Sent: Friday, February 23, 2007 1:40 PM Subject: Re: [Rd] Bootstrapping stepAIC() with glm.nb() > You did not say what the problem

Re: [Rd] Bootstrapping stepAIC() with glm.nb()

2007-02-23 Thread Prof Brian Ripley
You did not say what the problem was! But you are asking that an object which is not in scope (index) be found a few levels down. You should be able to fix this by substituting in the values in fn. Here is one way: up.obj <- update(object, data = data[index[, i], ]) Call <-

[Rd] Bootstrapping stepAIC() with glm.nb()

2007-02-23 Thread Dimitris Rizopoulos
Dear all, I would like to Boostrap the stepAIC() procedure from package MASS for variety of model objects, i.e., fn <- function(object, data, B = 2){ n <- nrow(data) res <- vector(mode = "list", length = B) index <- sample(n, n * B, replace = TRUE) dim(index) <- c(n, B) for (