Layman’s abstract for paper on estimation of residential radon concentration in Pennsylvania counties by data fusion

Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.

The article featured today is from Applied Stochastic Models in Business and Industry, with the full article now available to read on Early View here.

Zhang, X, Pyne, S, Kedem, B. Estimation of residential radon concentration in Pennsylvania counties by data fusion. Appl Stochastic Models Bus Ind. 2020; 1– 17.

Radon, a radioactive colorless and odorless gas, is naturally abundant in the soil. The World Health Organization (WHO) and the U.S. Environmental Protection Agency (EPA) have established radon as a leading cause of lung cancer among non-smokers. The problem is particularly acute in the U.S. state of Pennsylvania (PA) where about 40% of homes have residential radon levels above the EPA action guideline. Hence, from a public health, as well as real estate, point of view, it is important to understand the behavior of residential radon in each of PA’s counties. Given a county, its probability distribution of residential radon concentration can be obtained from a sample taken from the county itself. However, more precise distribution estimates can be obtained by fusing or combining the sample with samples from neighboring counties. A useful method of doing so is referred to as the density ratio model (DRM) where the corresponding probability densities are assumed to be related. Accordingly, the ratio of any pair of densities is related by a so called “tilt”, where, traditionally, the tilt is taken as xed regardless of the pair. It is shown how to extend the method using \variable tilts.” As a case in point, samples from Beaver county and its neighboring counties of Allegheny, Butler, Lawrence, and Washington, were fused in the estimation of exceedance or threshold probabilities in Beaver county. That is, it is shown how to estimate the probability that radon concentration in Beaver county exceeds any given threshold using data from Beaver county itself and, additionally, from its neighbors, using the density ratio model paradigm. The resulting con dence intervals of the threshold probabilities are relatively short.