You can get a bootstrapped version of V by using the saving() option of
bootstrap, use the dataset of replications and then calculate the covariance
matrix on the coefficients:
corr b_* , cov
or:
mat accum myV= b_* , deviation nocons
mat myV=myV/(`r(N)'-1)
The e(V) you get at the end of bootstrap is what's left over from the last
ressample.
Michael Blasnik
[email protected]
----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Sunday, June 20, 2004 5:08 PM
Subject: st: obtaining bootstrapped variance covariance matrix using version
6
> Hi. I am using the bstrap command in version 6 to obtain bootstrapped
> standard errors. However, I wish to obtain the entire bootstrapped
> variance covariance matrix. After the bstrap command I typed matrix V =
> e(V) and then matrix list V, but the matrix I get does not appear to be
the
> correct one (the square root of the variances in this matrix do not give
me
> the bootstrapped standard errors). I'd appreciate any advice on this.
> Thank you!
>
>
> Charlene M. Kalenkoski, Ph.D.
> Ohio University
> Department of Economics
> Bentley Annex 351
> Athens, OH 45701
>
> Phone: (740) 593-2022
> Fax: (740) 593-0181
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
