About: Hurricane Maria, what do we know more than five years later?
Basically, if you don't understand statistics and can't tell the difference between survey data and administrative data, preliminary data and revised data, don't act recklessly. Consider the image below.
figure 1: Cumulative excess deaths since September 2017, for simple time dummy OLS model (blue), population-adjusted OLS model (green), and population-adjusted quantile regression model (red), all based on the data in this report Methods postal, Milken Institute point estimate (black square) and 95% confidence interval (gray +), Santos-Lozada, Howard letter (yellow-green triangle), Cruz-Cano and Mead (pink square), Kopits “firm” maximum (cyan triangle). Light green band: Kopits estimate In October 2018 it was 300-400. Milken Institute (2018), Santos-Lozada and Howard (2018), Cruz-Cano and Mead (2019)and Kopit(2018a), Kopit(2018b).
From Sandberg et al. (July 2019) Epidemiology:
Mr. Kopitz's ire centered on Kishore's research. While the point estimate for the end of 2018 (4545) is higher than other estimates, the confidence interval includes the lower estimate. Mr. Kopits' provisional scope of 5/31 does not include other scopes. (must admit Mr. Kopits did not understand what a confidence interval was at the time.).
