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

Re: st: xtivreg, fd

From   "Mark Schaffer" <>
Subject   Re: st: xtivreg, fd
Date   Fri, 30 Apr 2004 17:00:04 +0100


Date sent:      	Fri, 30 Apr 2004 13:04:56 +0200
From:           	"Antonio Rodrigues Andres" <>
To:             	<>
Subject:        	st: xtivreg, fd
Send reply to:

> Dear Stata users
> I am trying a simple dynamic model with the lagged value of the dep.
> variable as regressor. I pretend to obtain the Anderson-Hsiao estimator
> using yit-2 as instrument but i did not succeed. I should get the same
> results using xtivreg, fd and ivreg with the differenced variables
> Does anybody knows why not
> sort country year
> by country: gen lsrt_1=lsrt[_n-1]
> by country:gen lsrt_2=lsrt[_n-2]

Does Stata know at this point that your data are panel data?  If not, 
some of the lags might be messed up.  For example, the 1-period lag 
for the first time period of observation 2 should be missing, but 
Stata might be using the latest time period of observation 1 instead. 
If you use tsset and then time series operators to create the lags, 
e.g., l2.lsrt instead of lsrt[_n-2], this won't happen and the lags 
will be correct.

Hope this helps.


> gen dlsrt=lsrt-lsrt_1
> gen dlsrtl=lsrt_1-lsrt_2
> Model yt=  alpha* yt-1 + eta _i +ut
> variables have been already differenced
> ivreg dlsrt (dlsrtl=lsrt_2) using yit-2 as instrument
> tsset country year
> xtivreg lsrt (lsrt_1=lsrt_2), i(country) fd
> Regards
> Antonio
> *
> *   For searches and help try:
> *
> *
> *

Prof. Mark E. Schaffer
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS  UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
*   For searches and help try:

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