Statistical Methods

This report contains two studies, by Drs. Brent A Coull and Eun Sug Park and their colleagues, and a Commentary discussing each study individually, as well as an Integrative Discussion of the two. 
Part 1. Statistical Learning Methods for the Effects of Multiple Air Pollution Constituents. Brent A. Coull et al.
Part 2. Development of Enhanced Statistical Methods for Assessing Health Effects Associated with an Unknown Number of Major Sources of Multiple Air Pollutants. Eun Sug Park et al. 

This report describes semiparametric methods for epidemiologic investigations of the short-term effects of air pollution on health, intended specifically to improve the reliability of point estimates and confidence intervals. Dr. James Robins of the Harvard School of Public Health and colleagues developed the new methods, used simulations to compare them with other methods, and applied them to a large epidemiologic data set from the National Morbidity, Mortality, and Air Pollution Study (NMMAPS) to assess their effectiveness. The report is accompanied by a short editorial to assist the reader in understanding this study and its contributions to epidemiologic methods for air pollution.