Panel data can be analyzed for average of multiple cross sectional data?

Question

stack users.

I have panel data (cross sectional at 3 different points in time) on my hand. What I want to do is to check some effects only at the cross-sectional level.

But as my data is originally collected for panel purpose, the samples are not random across the waves.

In this case, as far as I know I cannot pool the data. Could anyone help me understand the logic behind why the pool is not working? Is it because of the correlation between samples?

And would averaging of each estimation be advantageous over the pool data in my case?

Any valid reason or suggestion would be appreciated!

question from:https://stackoverflow.com/questions/65930038/panel-data-can-be-analyzed-for-average-of-multiple-cross-sectional-data

Answer 1

Waitting for answers