[R] boot function returns the same results every time - there appears to be not resampling of the original data.

[Peter Ehlers]

John,

  packageDescription('boot')

will give you a hint.

  -Peter Ehlers

John Sorkin wrote:
> Prof Ripley,
> Thank you for your advise. You recommend I consult the "book for which
> this is support software." I would be happy to do so, but I don't know
> what book you are referring to. 
> Thank you,
> John
> 
> 
> 
> 
> 
> John David Sorkin M.D., Ph.D.
> Chief, Biostatistics and Informatics
> University of Maryland School of Medicine Division of Gerontology
> Baltimore VA Medical Center
> 10 North Greene Street
> GRECC (BT/18/GR)
> Baltimore, MD 21201-1524
> (Phone) 410-605-7119
> (Fax) 410-605-7913 (Please call phone number above prior to faxing)>>>
> Prof Brian Ripley <ripley at stats.ox.ac.uk> 12/30/2009 8:55 AM >>>
> On Wed, 30 Dec 2009, John Sorkin wrote:
> 
>> R 2.8.1
> 
> Rather old: have you updated package boot since?
> 
>> windows XP
>>
>> I am trying to learn how to use the boot function to perform a 
>> bootstrap of a regression. I have written a short trial program, 
>> shown below. Clearly I have done something wrong as the output of 
>> each of the 100 bootstrap values for the regression are exactly the 
>> same - there does not appear to be any bootstrap respampling!. What 
>> have I done wrong?
> 
> First, sim="parametric" does not give 'bootstrap respampling', so I 
> wonder if you have consulted the book for which this is support 
> software?  You will find it hard to use boot without a clear grasp of 
> the underlying concepts.
> 
> For bootstrapping a regression (and why you almost certainly do not 
> want to do that) see the examples in MASS.
> 
> You clearly missed the following on the help page:
> 
>       For the parametric bootstrap it is necessary for the user to
>       specify how the resampling is to be conducted.  The best way of
>       accomplishing this is to specify the function ̔ran.gen̓ which
> will
>       return a simulated data set from the observed data set and a set
>       of parameter estimates specified in ̔mle̓.
> 
> The default ran.gen() function just returns the data ....
> 
>>
>> # Define function to be run. Function will return
>> # beta coefficeint for x.
>> fitter<-function(d)
>> {
>>  fit1<-lm(y~x,data=d)
>>  print(names(fit1))
>>  print(summary(fit1))
>>  summary(fit1)$coefficients[2,1]
>> }
>>
>> # Define dataframe
>> x<-1:10
>> y<-x+rnorm(10)
>> d<-data.frame(x,y)
>>
>>
>> #Run boot strap
>> boot(d,fitter,R=100,sim="parametric")
>>
>>
>>
>> John David Sorkin M.D., Ph.D.
>> Chief, Biostatistics and Informatics
>> University of Maryland School of Medicine Division of Gerontology
>> Baltimore VA Medical Center
>> 10 North Greene Street
>> GRECC (BT/18/GR)
>> Baltimore, MD 21201-1524
>> (Phone) 410-605-7119
>> (Fax) 410-605-7913 (Please call phone number above prior to faxing)
>>
>> Confidentiality Statement:
>> This email message, including any attachments, is for\...{{dropped:25}}
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

-- 
Peter Ehlers
University of Calgary
403.202.3921