The outcome variable in my dataset is cost. I prefer not to transform it as
readers will really be interested in pounds, not log(pounds) or
sqrt(pounds). I have fitted my model using lme and now wish to use boot on
it. I have therefore plagiarised the example in the article in RNews 2/3
Decembe
We don't have the structure of your dataset. But it seems pretty clear
that your resampling is not preserving the random effects structure: you
are always fitting to the same clusters labelled by gp and so missing the
major source of variability.
You either need to resample clusters, or resamp
Fra: [EMAIL PROTECTED] på vegne af Prof Brian Ripley
Sendt: ti 21-06-2005 18:53
Til: Michael Dewey
Cc: r-help-stat.math.ethz.ch
Emne: Re: [R] Problem trying to use boot and lme together
We don't have the structure of your dataset. But it seems pretty clear
that your resampling i
On 6/21/05, Søren Højsgaard <[EMAIL PROTECTED]> wrote:
> The problem with simulate.lme is that it only returns logL for a given model
> fitted to a simulated data set - not the simulated data set itself (which
> one might have expected a function with that name to do...). It would be nice
> wit
On Tue, 21 Jun 2005, Douglas Bates wrote:
On 6/21/05, Søren Højsgaard <[EMAIL PROTECTED]> wrote:
The problem with simulate.lme is that it only returns logL for a given model
fitted to a simulated data set - not the simulated data set itself (which one
might have expected a function with that
At 23:09 21/06/05, Prof Brian Ripley wrote:
>On Tue, 21 Jun 2005, Douglas Bates wrote:
>
>>On 6/21/05, Søren Højsgaard <[EMAIL PROTECTED]> wrote:
Thanks everyone for your help, more comments at the foot
>>>The problem with simulate.lme is that it only returns logL for a given
>>>model fitted to