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: unrecognized brant command

From   Richard Williams <>
To,, Nick Cox <>
Subject   Re: st: unrecognized brant command
Date   Mon, 12 Nov 2012 09:26:31 -0500

At 08:02 AM 11/12/2012, wrote:
Thanks Nick,
I managed to get the brant command from a lower version of the spost.

However, when i run the brant command I get the following error message.

. brant, detail
not all independent variables can be retained in all binary logits
brant test cannot be computed

What does this error message mean in a less technical language? When I
click on the r(999) I get a stata dialogue message which reads "no
entries found for search on "r(999)"."


The -brant- command takes your ordinal dependent variable, dichotomizes it in various ways, and then runs a series of logistic regressions. So, if your DV has 4 values, the first logistic regression is category 1 versus categories 2, 3, 4. Then, it does 1 & 2 versus 3 & 4, and finally, 1, 2, 3 vs 4. Your error message is telling you one or more of those logistic regressions can't be run without dropping some variables. Perhaps category 1 only has 7 cases and your model has 8 variables. My guess is you will need to simplify the model or combine categories. You can run the logistic regressions yourself to pinpoint exactly where and why the problem is occurring. But I would start just by doing a frequency of the DV to see if any categories (especially the first and the last) have very small frequency counts.

Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index