Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Stata RE Probit (about quadrature)

From   "Guillaume Frechette" <>
Subject   RE: st: Stata RE Probit (about quadrature)
Date   Tue, 01 Jun 2004 12:36:02 +0000

Glad I could help. To understand a bit more about quadrature points you would need to read on numerical integration. The book I have learned with is "An Introduction to Numerical Analysis" by Atkinson. Although I don't think it covers Guauss-Hermit (which is what is used by RE probit) it will explain the basic idea of numerical integration.



From: "Clive Nicholas" <>
Subject: RE: st: Stata RE Probit
Date: Sat, 29 May 2004 00:50:48 +0100 (BST)

Guillaume Frechette replied to Aparna Mathur:

> This is usually a convergence problem. You can try changing the number of
> quadrature points or install superxt
> ( which does that
> automatically as well as trying different starting values for rho.

Thanks for this. I'd had problems myself when fitting -xtprobit- models a
few months ago and wasn't aware of a solution. Perhaps now I can start to
understand at a practical level what quadrature points are!

CLIVE NICHOLAS |t: 0(044)191 222 5969
Politics |e:
Newcastle University |
* For searches and help try:
FREE pop-up blocking with the new MSN Toolbar get it now!

* For searches and help try:

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