Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: iv probit and endogenous binary variables

From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   RE: st: iv probit and endogenous binary variables
Date   Thu, 14 Sep 2006 12:19:46 +0100

Your question I see as a statistics question, 
modulo your use of econometric terminology, "endogenous"
and (tacitly) "instrumental variables". I am not aware of 
anything in your question that is intrinsically econometric. 

The list is declared in the FAQ to allow statistics 
questions. Many people would rather that others didn't 
allow questions that are very elementary or that are
easily answered by consulting standard texts, but 
yours doesn't appear to be either, so no problem there. 

However, there is general advice, often ignored, 

"Many questions will be of interest to only some Statalist 
members. Always recall that members come from many different 
sciences. Try to show a little sensitivity to those 
noneconometricians, nonbiostatisticians, or whoever it may 
be who may know little about your question. Flag your 
question as of limited interest, so that people can 
delete it quickly, or take the trouble to avoid 
subject-specific jargon." 

That's a counsel of perfection, and no one is suggesting, 
say, that anybody has to include a mini-tutorial on 
instrumental variables for the sake of those who 
are in blissful ignorance. 

That said, it is always a little puzzling when people
send questions that they themselves appear to think are 
inappropriate for the list. 

[email protected] 

Nishant Dass
> What should one do if one also has an interaction term with
> this endogenous binary variable on the RHS?  (I realise
> this  has little to do with Stata and is simply an
> econometrics question, so please excuse me for that.) 

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index