Wage draws from lognormal in dgp() function are sometimes way too large s.t. when we exponentiate we get infinity

[jamiefogel]

In bartik_analysis.py we drop obs with very large wages so this doesn't actually create an Error, just a RuntimeWarning. I'm just kinda stumped and ready to give up on this for now, especially since it's only going to matter in some of our secondary analyses.

https://github.com/jamiefogel/Networks/blob/39edb1f930ef5fea01304ecfc016c88e8f68527e/Code/Modules/dgp_func.py#L89C1-L91C5