Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: xtabond2 - How deep can the instrument go?
From
urbain thierry YOGO <[email protected]>
To
[email protected]
Subject
Re: st: xtabond2 - How deep can the instrument go?
Date
Tue, 8 May 2012 18:19:31 +0100
the point is that a rule of thumb in this matter is that the number of
instruments should not exceed the number of id (house).
2012/5/8, San K <[email protected]>:
> Dear all,
> My panel consists of 9000 houses X 22 quarterly metered consumption.
> Unfortunately I have lot of persistence in the data.
>
> After basing around different estimators I decided to settle with xtabond2.
>
> Here is my simplified semi-log model.
> xtabond2 y L4.y price L4.price L4.price L4.price L4.price WeatherVar
> SeasonalDummies, noleveleq
> gmmstyle(y, laglimits(16 16) equation(diff))
> ivstyle(PriceVariables, p)
> ivstyle((WeatherVariables, equation(diff))
> ivstyle(SeasonalDummies, equation(diff))
> twostep ar(5) robust
>
> I started my instrument from Lag5 but none of them passed the Hansen
> test or Arellano-Bond test for AR().
>
> My burning question is... is it okay to use the 16th lag as instrument
> given I have only 22 time steps?
>
> Regards,
> devank
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
--
*Urbain Thierry YOGO
P.h.D candidate in Economics*
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/