Eric Miller, a Consultant at Summit, recently submitted an article to the Stata Journal, a publication that highlights new techniques and applications pertaining to the Stata statistical software package. His draft of the article was years in the making, and discusses his implementation of an optimization method for selecting boundary points for a stratified random sample.
Stratified sampling is a statistical technique which can be used to reduce variability when selecting random samples of data. Eric's optimal stratification program is currently used on a number of Summit projects including enforcement analytics at the U.S. Department of Labor.
In addition to articles focusing directly on new applications of Stata, the Stata Journal includes discussions on statistics, teaching methods, and data analysis. Eric’s paper is currently under review for inclusion in a future issue. Until then, you can read the current edition at their website.