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: using the estimates of each individual bootstrap iteration


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: using the estimates of each individual bootstrap iteration
Date   Wed, 3 Aug 2011 22:19:49 -0400

Pavel Roshanov <roshanp@mcmaster.ca>:
Wrap the work in a -program- that:
loads the data (-use-)
resamples (-bsample-)
computes significance of each coef and returns a one or zero, meaning
sig or not,
clear estimation results (-ereturn clear-)
then -end-s.

Then loop with -forv- or -simulate- 10000 times.

See also
http://www.stata.com/statalist/archive/2006-12/msg00471.html

On Wed, Aug 3, 2011 at 9:19 PM, Pavel Roshanov <roshanp@mcmaster.ca> wrote:
> Hi Statalisters,
>
> I didn't receive a response and am still hoping for some help. Let me
> reiterate my problem:
> I want to test the same set of independent variables in a large number
> of bootstrap samples and determine the proportion of times that each
> variable came up statistically significant (p=<0.05). I suspect this
> is simple to do but I just can't figure it out!
>
> Thank you for your time
>
> Pavel Roshanov
> Graduate Student
> Department of Clinical Epidemiology and Biostatistics
> McMaster University, Hamilton, ON, Canada
>
> On Tue, Aug 2, 2011 at 2:16 PM, Pavel Roshanov <roshanp@mcmaster.ca> wrote:
>>
>> Dear Statalist,
>>
>> When running some estimation procedure with bootstrap resampling
>> (logistic regression in my case) is there some way to store/use the
>> results of each of, say, 10 000 iterations? I am trying to calculate
>> the proportion of times out of 10 000 that a coefficient comes up
>> statistically significant in order to test the stability of my
>> findings.
>>
>> Any help will be greatly appreciated.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


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