[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Bootstrap bias corrected p-values

From   "John Wald" <[email protected]>
To   <[email protected]>
Subject   st: Bootstrap bias corrected p-values
Date   Tue, 18 Sep 2007 11:03:24 -0500


Is there an easy way to get p-values from the bias corrected estimates
of a bootstrap?  I get the bias-corrected 95% confidence interval, but
it would be nice to get a p-value corresponding to whether the estimated
coefficient from my messy estimation procedure is <= 0.  Since the
distribution of the estimated coefficients is highly non-normal, looking
at the estimated standard errors isn't helpful.   I can save the
estimated betas and add the bias, but I believe that the bias corrected
estimate is better than what you get if you just look at the percentage
of non-positive estimated coefficients.

Thanks in advance,


John Wald
Department of Finance
College of Business
University of Texas at San Antonio
One UTSA Circle
San Antonio, TX  78249

[email protected]

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index