Generalized method of moments

[@loska23]

Excellent video! I've never found a good simple explanation of "moments" in statistics

[@mronkko]

If you just want to know what the term moment refers to, I think wikipedia provides a good explanation en.wikipedia.org/wiki/Moment_(mathematics)

[@kvafsu225]

Very good introduction. Thanks

[@mohammadkashem3375]

Dear Sir, Is there any way to use System GMM model in time series data? In one of paper an anonymous reviewer has suggested to use System GMM model for time series data. However, as per as I know System GMM can be used only for Panel data when there is endogeneity problem. Could you suggest me what should I do in this case?

[@mronkko]

I am not aware of any estimators that would require endogeneity to be present and system GMM certainly does not. What GMM allows you to do is to consistently estimate a model with lagged dependent variable as a control. Note that Arellano-Bond estimator and the System GMM are not general tools for addressing endogeneity but they address the spesific problem of dynamic panel bias.

[@user-wr4yl7tx3w]

this is really well explained and organized. i think i am finally getting it.

[@mronkko]

You are welcome!

[@diyanaathirahadnan8418]

I have problem. I cannot found the rule of thumb choosing between difference or system by Bond (2002). If someone happen to have the article, I really appreciate if you can share with me. Thank you

[@mronkko]

I would go for Roodman, D. (2009). How to do Xtabond2: An Introduction to Difference and System GMM in Stata. The Stata Journal, 9(1), 86-136. doi.org/10.1177/1536867X0900900106 Use the system estimator of you think that the additional assumptions made about the instruments are justified.

[@diyanaathirahadnan8418]

@@mronkko Thank you very much Sir.

[@nguyenthile2796]

Thank you so much 17/07/2022

[@mronkko]

You're most welcome

[@propername4830]

Got here from "gmm moments"

[@mronkko]

Wow. Google seems to like my video ;)

[@aanarief6896]

thank you

[@mronkko]

Welcome!