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, is already up and running.

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

Re: st: Generating dummies with xi

From   Nick Cox <>
Subject   Re: st: Generating dummies with xi
Date   Tue, 12 Apr 2011 08:36:43 +0100

Good point. My statement was an oversimplification. It very much needs
a qualification: when there is a choice.


On Tue, Apr 12, 2011 at 2:37 AM, Garry Anderson
<> wrote:

> I think another exception to use -xi:- with Stata 11, rather than use
> factor variables, is when the following estimation commands are being
> used
> -exlogistic-, -expoisson-, -qreg-, -iqreg-, -sqreg- and -bsqreg-.
> There may be other commands that do not allow factor variables.
> Best wishes, Garry
> -----Original Message-----
> From:
> [] On Behalf Of Nick Cox
> Sent: Monday, 11 April 2011 10:51 PM
> To:
> Subject: Re: st: Generating dummies with xi
> It is better to use factor variables directly than to use -xi:-.
> The only exception would be if are using an out-of-date version of
> Stata, say Stata 10, in which case you are asked to tell us that in the
> Statalist FAQ.
> That said, there seems to be some confusion here. First, -xi- will
> generate all pertinent dummies if asked directly. Second, a modelling
> command necessarily omits at least one such variable; that's the way
> that dummies work. Third, the help for -xi:- explains the use of
> -char- to get what you want, at least in Stata 10. (I don't have access
> to earlier versions at this time.)
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index