Came back to see this again. Learning a lot. Thanks so much John Antonakis. To Amar Anwar: he dramatizes in the first 10 minutes or so, which is why we all like this video so much. Then John gets down to serious business. See the second half of his podcast or see his "For researchers" version if the podcast. If you know all that stuff well then good for you. But I think that most students in a non economics degree or those learning econometics will learn a lot from this video.
Dear Prof. John Antonakis, Thank you so much for providing new directions in solving the common-method problem. This is very helpful. If possible, would you please prepare a video regarding how to solve the common-method variance problem with 2SLS using SPSS or AMOS? Thank you. Mike
THANK YOOOOOOU! Thank you so much for this great video! People like you make the world a better, more educated place! You helped me so much with my assignment. :-)
His presentation is somewhat confusing, and he switched what U is from time to time. U is just an error term like e. Sometimes he says that U is the residual in y=f(x)+u, and sometimes in x=f(z)+u. Similarly for E... sometimes the residual in x=f(y)+e, and sometimes in X=f(z)+e. See my video called Mailbag: Notation for more about U's and E's in general in econometric notation.
Dear Professor! Thank you very much indeed for sharing this great video :) I am wondering if you could share a video on how to apply the concepts you have discussed in this video, preferably from the paper you mentioned in this video. The paper is a good read and useful. I feel truly grateful to you and would appreciate your kind response :)
can someone please explain what the "U" term is? I understand the e term as all the movements in Y that X did not predict, but am having trouble understand what U is and thus how U and e would be correlated
in other words, if there is a perfect 1:1 relationship btw riflefired and soundheard, then how does riflefired explain diskshatter better than soundheard? i understand that the reality is that riflefired physically causes soundheard but why should it matter statistically, since riflefired isn't adding anything new to explaining diskshattered? if the omitted variable doesn't explain the depended variable any better than the endogenous one is it ok to forget about it?
Dear professor, while your paper is great, you use notation β1 = inconsistent. This has created some confusion, because a parameter is not consistent or inconsistent. β1 is referring to the estimator, which is an odd notation. Thanks anyway.
Thanks Dr. Antonakis. Would be amazing to see a follow up to discuss issues that you briefly touched on, such as endogeneity in HLM. I wish more people (including myself) truly and completely understand the content you are delivering. Link to his paper for further reading: datascienceassn.org/sites/default/files/On%20making%20causal%20claims%20A%20review%20and%20recommendations.pdf
A paper will be out shortly just on this topic in a couple of months or so (look out for it here: scholar.google.de/citations?hl=fr&user=nWTsugIAAAAJ&view_op=list_works&sortby=pubdate
Great explanation! (Except for talking about Swiss Francs with Euros bills falling in the backgroud. We are not in the Euro zone, and proud not to be!)
Endogeneity is bad. I am looking for engineering examples of these things. Like stress and strain for example in tension test using the language of statistics/econometrics. Dunno if that's been done.
This is an extremely helpful (and entertaining) introduction to endogeneity that will be useful for scholars operating in many disciplines.
Amazing explanation. I knew nothing about it, now i understood it very well. Thank you
Finally, a comprehesible video on endogeneity. Thank you for this really excellent presentatoin.
The nicest explanation ever!!!! I also learned how to make a presentation and make complex issues to look simpler!!
This is the best Video I have ever watched on econometrics. The professor is really fun to watch! His accent keeps me awake
Hats off to you professor! You made the concept very simple. Thanks very much.
Came back to see this again. Learning a lot. Thanks so much John Antonakis. To Amar Anwar: he dramatizes in the first 10 minutes or so, which is why we all like this video so much. Then John gets down to serious business. See the second half of his podcast or see his "For researchers" version if the podcast. If you know all that stuff well then good for you. But I think that most students in a non economics degree or those learning econometics will learn a lot from this video.
Very engaging and elucidating video. The graphics were a wonderful extension of your explanation making the concepts very clear.
Thank you for this luminous explanation. Teaching is good but being educational is better.....
great presentation with clear explanation and clever animations. Thanks for posting this video for free!!
Dear Prof. John Antonakis,
Thank you so much for providing new directions in solving the common-method problem. This is very helpful.
If possible, would you please prepare a video regarding how to solve the common-method variance problem with 2SLS using SPSS or AMOS? Thank you. Mike
Thanks for making this video. Really clear explaination of endogeneity and the use of TSLS.
This is an excellent video!
Perfectly explained ! Thanks for your efforts
The best econ-video ever! Thank you so much!
Hello from Prague, Charles University! Thanks a lot for this video! It will definitely save me on the exam today )))
Great explanation of endogeneity concept; thank you professor
Great insight into common errors made when modelling phenomena, cheers from Canada
Loved the presentation... Excellent resume...
Instructive, thanks!
THANK YOOOOOOU! Thank you so much for this great video!
People like you make the world a better, more educated place! You helped me so much with my assignment. :-)
very nice....explained in the most easiest way.thanks sir
Thank you for this great vid! It changed my naive usage of regression analysis. In the future I will most certainly look fpr endogeneity
Amazing presentation! Thank you very much!
Bravo! Very clear explanation!
Excellent video!
Thank you so much Prof.
Brilliant and intuitve!
His presentation is somewhat confusing, and he switched what U is from time to time. U is just an error term like e. Sometimes he says that U is the residual in y=f(x)+u, and sometimes in x=f(z)+u. Similarly for E... sometimes the residual in x=f(y)+e, and sometimes in X=f(z)+e. See my video called Mailbag: Notation for more about U's and E's in general in econometric notation.
Dear Professor! Thank you very much indeed for sharing this great video :) I am wondering if you could share a video on how to apply the concepts you have discussed in this video, preferably from the paper you mentioned in this video. The paper is a good read and useful. I feel truly grateful to you and would appreciate your kind response :)
Helpful! Thank you!
Thank you! Very interesting.
very well explained. thx sir
love how you explained randomization tho
Thank you - this was great
Thanks. It is very helpful.
Thank you
i created gmail account specially to post this:
THANK YOU!
POLI 210 Mccill whats good fam
can someone please explain what the "U" term is? I understand the e term as all the movements in Y that X did not predict, but am having trouble understand what U is and thus how U and e would be correlated
in other words, if there is a perfect 1:1 relationship btw riflefired and soundheard, then how does riflefired explain diskshatter better than soundheard? i understand that the reality is that riflefired physically causes soundheard but why should it matter statistically, since riflefired isn't adding anything new to explaining diskshattered? if the omitted variable doesn't explain the depended variable any better than the endogenous one is it ok to forget about it?
Dear professor, while your paper is great, you use notation β1 = inconsistent. This has created some confusion, because a parameter is not consistent or inconsistent. β1 is referring to the estimator, which is an odd notation. Thanks anyway.
Thanks Dr. Antonakis. Would be amazing to see a follow up to discuss issues that you briefly touched on, such as endogeneity in HLM.
I wish more people (including myself) truly and completely understand the content you are delivering.
Link to his paper for further reading:
datascienceassn.org/sites/default/files/On%20making%20causal%20claims%20A%20review%20and%20recommendations.pdf
A paper will be out shortly just on this topic in a couple of months or so (look out for it here: scholar.google.de/citations?hl=fr&user=nWTsugIAAAAJ&view_op=list_works&sortby=pubdate
Hello, how can i use 2sls in SPSS? Please help!!
Great explanation! (Except for talking about Swiss Francs with Euros bills falling in the backgroud. We are not in the Euro zone, and proud not to be!)
I was able to understand somewhat but it is a little out of my scope. :)
prof i disagree with you saying there is no corellation. there is corellation but there is no causation.
Endogeneity is bad. I am looking for engineering examples of these things. Like stress and strain for example in tension test using the language of statistics/econometrics. Dunno if that's been done.
Can we do endogenity test in eveiws?
Hello Mister Yianni!
Great session, your email
More Drama less knowledge.