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: how to write model equation for probit with cluster

From   Maarten buis <>
Subject   Re: st: how to write model equation for probit with cluster
Date   Sat, 12 Feb 2011 18:02:30 +0000 (GMT)

--- On Sat, 12/2/11, Laura R. wrote:
> I estimated a probit equation in stata with the option
> -cluster-, like this:
> probit depvar indepvar1 indepvar2 ..., cluster(clustervar)
> Usually, you can write the formula for probit models in the
> paper something like this:
> Pr(Y=1|observed var)= "cdf sign" (bX1+bX2+e)
> However, I am wondering how to insert the option for
> cluster into this formula.

You should not include the e (I assume error term) in these
models. The error terms is already included in the probability.

Compare this with linear regression, either you say:

y = bx1 + bx2 + e


E(y) = bx1 + bx2

but not:

E(y) = bx1 + bx2 + e

Other than that, you can use the same formula for model with
clustered standard errors, as this formula only represents
the relationship between the mean/probabilty and the explanatory

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

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