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 on April 23, and its replacement, statalist.org is already up and running.


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

st: bootstrap p-value and predictions


From   Shehzad Ali <drshehzad_ali@yahoo.com>
To   Stata List <statalist@hsphsun2.harvard.edu>
Subject   st: bootstrap p-value and predictions
Date   Fri, 18 Nov 2011 12:32:04 +0000 (GMT)

Hi listers,

I have a basic question regarding bootstrap.


For instance, you are running linear regression with the -bootstrap- procedure to estimate bootstrap standard errors and p-values. In that process you also predict outcome for each bootstrap
sample using actual values of covariates but varying only the treatment
variable (setting equal to 0 and then 1) to get the difference between treatment 1 and treatment 0 for each bootstrap sample. After say 10,000 iterations, you get the bootstrap SE and p-values and also 10,000 predictions for the difference between treatment 1 and treatment 0. 


My question is: based on the SE and p-value for the treatment parameter, would you be able to estimate the proportion of the predictions from the bootstrap samples that should be above and below zero for the difference between outcomes in the treatments?


Thanks

Shehzad

*
*   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