Page 1 of 1
2SLS Manual Input vs. Eviews
Posted: Thu Oct 22, 2015 8:12 pm
by Amaziah
I am trying to test for simultaneity by estimating a 2SLS instrumental variable procedure. I estimated the two steps manually and then compared them to the Eviews preprogrammed 2SLS estimation and got extremely different results. I am obviously making a mistake in one, if not both, of the procedures. Can someone help correct my estimation mistakes?
Eviews Estimation
P (price) is the potential endogenous explanatory variable and othrsrvcs is the instrumental variable.
Equation Specification:
Code: Select all
att c expand rex p rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
Instrument List:
Code: Select all
expand rex othrsrvcs rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
- Eviews Estimation.png (41.93 KiB) Viewed 8794 times
Manual Estimation
I first regressed the potential simultaneous variable (p) on all of the other explanatory variables and the instrument (othrsrvcs).
Code: Select all
ls p c expand rex othrsrvcs rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
I saved the residuals as the series pfit. I then estimated the original specification with the pfit series instead of p:
Code: Select all
ls att c expand rex pfit rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
- Manual Estimation.png (38.36 KiB) Viewed 8794 times
Re: 2SLS Manual Input vs. Eviews
Posted: Thu Oct 22, 2015 8:51 pm
by startz
The problem is probably that you want the fitted values--not the residuals.
Re: 2SLS Manual Input vs. Eviews
Posted: Thu Oct 22, 2015 9:26 pm
by Amaziah
Thank you. I want to include both the residuals and fitted value of the possible endogenous variable correct?
I generated the fitted values by forecasting the historical observations (phat). I estimated the following:
Code: Select all
ls att c expand rex (phat+pfit) rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
- Manual Estimation.png (41.87 KiB) Viewed 8781 times
Eviews Estimation
- Eviews Estimation.png (41.97 KiB) Viewed 8781 times
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 6:19 am
by startz
No.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 6:38 am
by Amaziah
startz wrote:No.
You are referring to including the residuals and the fitted values of the endoeneous variable? I got that from pages 6 and 7 from following source:
http://www.nyu.edu/classes/nagler/quant ... tes_oh.pdf. Is this methodology incorrect or just different?
But below are the results only using the fitted values of the endogenous variable. Is the discrepancy between methods a result of a mistake in my use of the Eviews 2SLS method?
Code: Select all
ls att c expand rex phat rinc pop ur win oppwin hwin prev rank temp precip night hist comm finale hc conf lastgame gamescoached rtv ntv
- Manual Estimation.png (36.22 KiB) Viewed 8764 times
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:05 am
by startz
1. What is the discrepancy you are seeing?
2. The link you posted says to use the fitted values.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:15 am
by Amaziah
startz wrote:1. What is the discrepancy you are seeing?
For example, when I use the preprogrammed 2SLS on Eviews I get the following results, which are much different than the ones I just posted.
- Eviews Estimation.png (41.74 KiB) Viewed 8760 times
startz wrote: 2. The link you posted says to use the fitted values.
It says that X=X^+v^.
And then says to substitute X^ into the original equation for X: Y=a+b1(X^+v^)+b2X2+u.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:21 am
by startz
It says substitute in X^.
Which results are different?
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:27 am
by Amaziah
startz wrote:Which results are different?
The standard errors, and therefore t-Statistics, and the summary statistics differ.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:31 am
by startz
That's right. Doing the two stages manually requires a correction to the standard errors, as the document you posted states.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 7:38 am
by Amaziah
startz wrote:Doing the two stages manually requires a correction to the standard errors.
How can I correct the standard errors?
Since the Eviews 2SLS estimation results are poor, this would suggest that X^ is a noisy predictor of X?
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 8:32 am
by startz
If by "poor" you mean low t-statistics, yes that could be due to a noisy predictor.
Re: 2SLS Manual Input vs. Eviews
Posted: Fri Oct 23, 2015 8:36 am
by Amaziah
Okay. Thank you very much for all your help and for walking me through this step by step.