Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Saving bootstrap samples


From   Margaret MacDougall <Margaret.MacDougall@ed.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Saving bootstrap samples
Date   Mon, 17 Jun 2013 22:05:19 +0100

Dear Nick

I am sorry if I did not provide all of the information you required. Under your item 1., your third option looks attractive. I have three variables in the one spreadsheet from which the bootstrap samples are to be taken and would favour generating bootstrap samples of these which are saved within the original spreadsheet, if possible. It would be ideal if the output requested for each of these samples was also stored in the original spreadsheet, in this case, alongside each corresponding sample, in one or more columns if possible.

Best wishes

Margaret

On 15/06/2013 09:16, Nick Cox wrote:
Apart from the word "saving", what you seek is unclear.

Some precision on exactly what you seek would help:

1. Different bootstrap samples as

each a separate dataset
each a block of observations in  a combined dataset
each a block of variables in a combined dataset

2. Results of computations on each

stored precisely how.

Nick
njcoxstata@gmail.com

On 15 June 2013 08:39, Margaret MacDougall <Margaret.MacDougall@ed.ac.uk> wrote:

Thanks for your reply. It is the saving of the bootstrap samples I was
specifically asking about, but I would wish to save them in addition to the
output generated for particular parameters. Is it possible to combine these
two tasks?
On 14/06/2013 14:40, Maarten Buis wrote:

On Fri, Jun 14, 2013 at 3:31 PM, Margaret MacDougall wrote:
I would value some tips on how best to go about saving bootstrap samples
generated by means of the bootstrap command in Stata.
Do you want to save the samples or the statistics computed in each
sample? The latter can easily be achieved using the -saving()- option.
The former is harder. You'll probably have to abandon -bootstrap- and
code the loop yourself and draw random observation in each iteration
using -bsample-. It is probably most convient to store those samples
as fweights, like so:

forvalues i = 1/10 {
      qui gen w`i' = .
      bsample , weight(w`i')
}

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/



--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index