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]

Re: st: Question About Using Permute Command

From   Nick Cox <>
Subject   Re: st: Question About Using Permute Command
Date   Wed, 10 Oct 2012 15:08:45 +0100

Slightly different results are what you expect with -permute- unless
you -set seed #- to the same # before each run.


On Wed, Oct 10, 2012 at 2:48 PM, Michael Horowitz <> wrote:
> I am interested in using the permute command to re-randomize an IV of
> interest as a robustness test. The materials on it seem a bit sparse,
> however, and I was wondering if someone could assist with the correct
> syntax. Here are the two possibilities I have come up with:
> 1. permute IV (the one I am randomizing) "regress DV IV control1 control2
> control3" _b, reps(10000)
> 2. permute IV (the one I am randomizing) _b, reps(10000) saving(permtest1,
> replace): regress DV IV control1 control2 control3
> Both seem to run, but they produce slightly different results and I am not
> sure which is right. I know people are busy, but I would appreciate any
> insight on which syntax is correct - or if both are wrong, what to use
> instead.
*   For searches and help try:

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