Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: bootstrap and derived dummy variable

From   Erasmo Giambona <>
To   statalist <>
Subject   st: bootstrap and derived dummy variable
Date   Wed, 15 Feb 2012 14:02:30 +0100

Dear All,

I can see why one would bootstrap the standard errors if a variable
estimated from a regression is for instance used as regressor. Would
things be the same if instead of using the estimated variable directly
in the regression, one derive a dummy variable from the estimated
variable and the dummy variable is then used in the "second-stage"

I would appreciate any thoughts and references on this.


*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index