Endogeneity between dependent and indepentent variable?

[Anela Kien]

Hello,
I have more of a general question regarding statistical terms as I want to do a regression analysis and I am familiar with the term "endogeneity". (relationship between error term and independent variable).

I read the following in a paper:
We avoid endogeneity concerns by adjusting the variable x used in our main analysis as explanatory variable for the variation in the dependent variable. Since the variable x was determined based on the values of the dependent variable, we thus eliminate the individual contribution of the firm to the calculated variable x. The firm-level aggregate does not therefore depend on the respective firm itself.

Firm in this case is the object of investigation.

I do not understand why this is a problem and why it is called "endogeneity". As I understand this term, it is differently used than in this case.

Would this also be a problem

I would be very grateful if someone could explain this to me!

Thanks a lot,
Anela

[George Ford]

Just cause someone says something doesn't make it legit.

Sounds like they used firm fixed effects in ivreg. ?

[Anela Kien]

Yes, they used firm and time fixed effects. How do you know and is that important for answering the question? So do you think, that endogeneity is not the case here or is the problem a different one or is there no problem? As it is from a peer rewied paper I was curious.

[George Ford]

The authors said it was endogenous and attempted to fix it. Whether or not it is important is another question and there are tests for erogeneity. The paper may have more detail about their thoughts on it.