Expected working paper by October 2019
I posit that the hourly variation of air pollution affects the probability of crime. To understand the connection between hourly changes in pollution and crime, I use administrative crime records from Los Angeles and New York police departments alongside monitoring stations data from the environmental protection agency. I infer causality through the use of ordinary least squares, Poisson, and probit fixed-effect models controlling for the census tract where the crime occurs, the hour, the month, and the year of the crime plus additional control covariates at the crime, tract, and city levels. Preliminary results point towards a large effect of ozone, fine particle matter, and carbon monoxide on criminality..
- Submission Status: Working paper expected by October 2019.