Two Stage Least Squares - example

[@SeeyaMahendru]

i have been struggling with this since the past hour. thankyou so much i understood this in less than 10 mins!!

[@RogerKlauser]

That explanation at the end was money. Definitely going to help me w/ my econometrics final this Sat!

[@brandonjacobs5452]

Ben, brother, you are saving my life right now

[@Byc845]

You explain the concept so clearly! Thx a lot!

[@olivier306]

This belongs in a museum

[@patiencedzigbordiakosuakor5171]

Thanks a lot. This is going to help me a lot in my current thesis.

[@AnevaSee]

Thanks. great ideas are explained with simple words, perfect!

[@filippopovic8514]

Dear all,
I would really appreciate if someone could explain me why does Ben include SAT variable in estimating CA in first stage of 2SLS. Aren't distance and quality of public transportation enough to estimate CA.
Thanks

[@MrThefonzzi]

Yeah... why does he include SAT in CA? Would you get highly correlated estimates in the new regression? i.e. beta1 and beta2 being highly correlated...

[@SomethingSoOriginal]

I also would like to know this

[@jzureich2121]

Filip Popovic I have been wondering the same thing. I think it may have something to do with the fact we have already established/assumed it has a 0 covariance with the error term in which case it is beneficial to include in the first stage of regression.

[@fjosh457]

it is a significant variable for the data generation process of scores. That reflects previous knowledge and background of some particular topic. even tough attendance for some particular students is zero; then scores would be influenced by IQ or any other training in the past.

[@vladrosca6785]

Thank you very much, Ben!

[@ireneisme8747]

But how can we know INTEREST is in the error term, what if DISTANCE, PTRANS are also in the error term?

[@mm22sapphire50]

hi! can you use the lagged value of the endogenous variable as an instrument using 2SLS model

[@patiencedzigbordiakosuakor5171]

Also interested to know

[@rudraguha4618]

Thank you so much for your work.Helped a lot :)

[@3foss191]

Thks grand Prof.

[@Noelson]

If you regress CA on SAT and IVs and find out that the coefficient of SAT is significant, isn't that suggesting multicollinearity which should have been avoided?

[@SomethingSoOriginal]

+Yuchen Li Yeah that's what i was wondering to, why SAT is included in the first-stage

[@yanfengli4821]

Yes, I am wondering too. Can anyone explain why include SAT in both first and second stage? I think in the second stage, CA hat and SAT will have multicolinearity problem.

[@62294838]

This is sooooo good an explanation!!

[@dhruvsadarangani755]

thanks love your videos

[@karlkani9]

Very Helpful, thank you very much !!

[@lemonsqueezy09]

Wow when you say "Exogenous " contra "Endogenous", they sound really familiar.

[@njungohfozaopascaline6277]

Thanks very much. It helped

[@Lukas-md3bk]

You rock dude!!

[@SomethingSoOriginal]

Thanks, very helpful

[@sarahyuan8635]

Thank you

[@faithlangat2510]

awesome