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: RE: catzero

From   Margaret MacDougall <[email protected]>
To   [email protected]
Subject   Re: st: RE: catzero
Date   Fri, 02 Mar 2012 09:06:57 +0000

Dear Nick

This is a courtesy message to say thanks for replying to my original message below and that I had made some further headway on discovering that the hard copy documentation I had on my shelf for Stata Version 9.0 is outdated in relation to how to use the mfp command. I have since opted to stick to the pdf documentation provided with Version 12.0.

Best wishes

On 08/02/2012 17:25, Nick Cox wrote:
-catzero()- is an _option_ of -mfp- (in my ignorance I tried to find a -catzero- command).

There is guidance on pp.1157-8 of the Stata 12 manuals.

[email protected]

Margaret MacDougall

I would welcome examples of the using of the command 'catzero'
specifically to request indicator variables for a subset of the
predictor variables within the context of using both the mfp command and
the sunsidiary logit and xpower commands. I understand the syntax for
using catzero itself (i.e. 'catzero(varlist') but would value seeing an
illustration of how best to place this syntax in relation to the other
commands above.  I also want the mfp procedure to understand that my
caegorical predictor variables are to be left untransformed while also
being included in the model along with the continous variables.

*   For searches and help try:

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

*   For searches and help try:

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