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:glm with bin family and link probit VS. probit

From   Richard Williams <>
Subject   Re: st:glm with bin family and link probit VS. probit
Date   Tue, 31 May 2011 20:37:50 -0500

At 06:53 PM 5/31/2011, Judy You wrote:
Dear Richard:

Thanks for your replying my question and your trusting probit over glm
in this case.

The reason that i run the two models is to check if they got the same
answer, as most of reference books said so. Now, we could see the
difference. The difference is even bigger when the marginal effect is
estimated after the modellings. Could any other experts explain if the
glm program could be improved?

Most of the time you will get the same results with glm and probit, e.g.

sysuse auto
probit foreign weight mpg
glm foreign weight mpg, link(probit) family(binomial)

You just happened to pick a problematic example because failure was predicted perfectly in some cases. -probit- apparently has the specialized code to deal with such cases while the jack of all trades -glm- does not. Also, keep in mind that your -glm- run never converged while -probit- did, which may further contribute to the seeming differences between the two.

In general, if you have a choice between glm and a more specialized program, you will usually want the specialist. The code will run more quickly, the post-estimation commands will be more appropriate and diverse, and the error-checking may be better.

I suppose a programmer could try to improve -glm-, but that could also make it more bloated, and may not really be worth it given that you already have -probit-.

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 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index