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
Eviews Estimation.png (41.93 KiB) Viewed 3316 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
Manual Estimation.png (38.36 KiB) Viewed 3316 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
Manual Estimation.png (41.87 KiB) Viewed 3303 times



Eviews Estimation

Eviews Estimation.png
Eviews Estimation.png (41.97 KiB) Viewed 3303 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
Manual Estimation.png (36.22 KiB) Viewed 3286 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
Eviews Estimation.png (41.74 KiB) Viewed 3282 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.