Instrumental variable regression

[Michelle Voskanyan]

I am working with a dataset collected from 40 different villages. I am trying to predict financial inclusion levels (a score from 0 to 12) based on the financial literacy score (0 to 5). Since I have an endogeneity problem, I am planning to use two instrumental variables. I am confused about whether I should use ivreg2 or ivreghdfe. Which IV regression option would be the best for my case? Additionally, how do I ensure that my instruments are valid, not weak, not under-identified, and that there are no over-identification issues? How do I know that the choice of the two instrumental variables are valid? I don't quite understand how to interpret the results of the tests I get from using ivreg2 (Hansen J, etc.). I would really appreciate it if someone could please help me.

[George Ford]

ivreg2 and ivreghdfe are comparable, though the latter allows for multiple fixed effects.

Both will give you a bunch of test statistics on the instruments. I recommend you figure out what each one is for and what is the null, rather than have someone tell you. You'll get more out of it.

[George Ford]

This describes all the statistics reported in detail.

https://journals.sagepub.com/doi/pdf/10.1177/1536867X0800700402

[Michelle Voskanyan]

Thank you. I appreciate your help.