I was wondering if I understand these correctly. Would an example of a stochastic regressor be weather? so when thinking about the sampling distribtuion and causality, I would think of repeated experiments where in each 'draw' the weather can be different? and thus no systematic correlation of the weather and other unobseravbles should be present?

But then if I think of using something like, the characteristics of land (for example, Kearney and Wilson (2017) use the presence of shale play under your county to exploit fracking production), this is a deterministic regressor, correct? In which case the repeated sampling experiment to think of when thinking of endogeneity would consist of the same units (counties) always being assigned the same level of treatment, but then whether the populations/other factors about those places are systematically correlated with the fixed characteristic?

Is that interpretation correct about stochastic vs nonstochastic regressors generally? It seems that the conceptual experiment of repeated sampling for these two different scenarios are wholly different, where the former is one where different units could hypothetically receive different levels of treatment in another draw, and the latter they are fixed.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.