disputing a fixed effects model assumption


Ninja say what!?!
Hi guys,

I'm in the middle of responding to a reviewer's claim on fixed effects models and wanted to reach out to see if any of you had any thoughts.

I used a fixed effects model to look at the effect of prenatal adversity (measured at birth and collapsed to three categorical groups) on ANS outcome in children from ages 6 months to 5 years. In order to get around the requirement of having multiple different measures within each subject, I crossed age (at which the outcome was measured) with the adversity.

What the reviewer says is that the goal of a fixed effects approach is to examine the effect of some experience on an outcome, controlling for baseline levels of the outcome. They claim that an assumption is that the experience being examined does not influence the baseline level of functioning on the outcome being examined.

I disagree with this. An example I am providing is from econometrics, where the effect of price on quantity demands is analyzed using fixed effects with the city as the subject.

Does anyone have comments or questions? Clarifications I should add?

Thanks a lot!