Page 1 of 1
Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 7:19 am
by IanKennedy1
Hi,
I am fairly new to eviews 8 software and i'm struggling in figuring out my IV 2SLS regression.
What I am trying to do is to measure patents (innovation)(independent) on the unemployment rate(dependent), while controlling various other variables such as inflation, interest rate, trade, FDI etc.
What would I have to enter into Eviews into the equation specification and instrument list.
Assistance would be greatly appreciated.
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 9:58 am
by IanKennedy1
bump
please
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 10:57 am
by startz
Have you looked for 2sls in the Help system?
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 11:08 am
by IanKennedy1
Hi, I am fairly new to eviews and dont really understand the wording, but yes I have, I understand simultaneous but generally on all the examples the second equation is a independent variable on the RHS but my control variables are controlling the left hand side.
What would I have to put into eviews if I wanted to show the effect of patents on unemployment taking into account the control variables?
Would it be,
Estimation Specification
unemployment_rate c patents
Instrument List
c FDI inflation interest tax_wedge trade union_density......?
or do i treat patents as a instrument?
Thanks for the help, very stressed atm
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 11:11 am
by startz
Hi, I am fairly new to eviews and dont really understand the wording, but yes I have, I understand simultaneous but generally on all the examples the second equation is a independent variable on the RHS but my control variables are controlling the left hand side.
What would I have to put into eviews if I wanted to show the effect of patents on unemployment taking into account the control variables?
Would it be,
Estimation Specification
unemployment_rate c patents
Instrument List
c FDI inflation interest tax_wedge trade union_density......?
or do i treat patents as a instrument?
Thanks for the help, very stressed atm
Your control variables need to be in the Estimation Specification. Then the instrument list is all the exogenous variables, that is those variables that are not affected by unemployment.
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 11:31 am
by IanKennedy1
I dont have exogenous variables, all I have is the unemployment rate (dependent), and my patents(variable that im examining), trade, tax_wedge, union_density, real interest rate.
Feel like i'm being a complete idiot.
Re: Two-Stage Least Squares Regression
Posted: Mon Mar 23, 2015 11:36 am
by startz
You're not being an idiot. Everyone's a beginner initially.
Your questions really have nothing to do with EViews. You might want to go back to your econometrics text and review when least squares can be used and when 2SLS is needed.