Patrick Ball | Searcharxiv

arXiv2020

Estimating the number of SARS-CoV-2 infections and the impact of social distancing in the United States

Understanding the number of individuals who have been infected with the novel coronavirus SARS-CoV-2, and the extent to which social distancing policies have been effective at limiting its spread, are critical for effective policy going forward. Here we present estimates of the extent to which confirmed cases in the United States undercount the true number of infections, and analyze how effective social distancing measures have been at mitigating or suppressing the virus. Our analysis uses a Bayesian model of COVID-19 fatalities with a likelihood based on an underlying differential equation model of the epidemic. We provide analysis for four states with significant epidemics: California, Florida, New York, and Washington. Our short-term forecasts suggest that these states may be following somewhat different trajectories for growth of the number of cases and fatalities.

stat.AP↗

arXiv2019

Estimating the Number of Fatal Victims of the Peruvian Internal Armed Conflict, 1980-2000: an application of modern multi-list Capture-Recapture techniques

We estimate the number of fatal victims of the Peruvian internal armed conflict between 1980-2000 using stratified seven-list Capture-Recapture methods based on Dirichlet process mixtures, which we extend to accommodate incomplete stratification information. We use matched data from six sources, originally analyzed by the Peruvian Truth and Reconciliation Commission in 2003, together with a new large dataset, originally published in 2006 by the Peruvian government. We deal with missing stratification labels by developing a general framework and estimation methods based on MCMC sampling for jointly fitting generic Bayesian Capture-Recapture models and the missing labels. Through a detailed exploration driven by domain-knowledge, modeling and refining, with special precautions to avoid cherry-picking of results, we arrive to a conservative posterior estimate of 58,234 (CI95% = [56,741, 61,289]), and a more liberal estimate of 65,958 (CI95% = [61,462, 75,387]) fatal victims. We also determine that the Shining Path guerrillas killed more people than the Peruvian armed forces. We additionally explore and discuss estimates based on log-linear modeling and multiple-imputation. We finish by discussing several lessons learned about the use of Capture-Recapture methods for estimating casualties in conflicts.

stat.AP↗