A Efficient Estimation of Generalized Partially Linear Single-Index Models Using Penalized Splines

Paper link : (working in progress)
Github link : (working in progress)
Description: Dr. Minggen Lu (First Author); Emmanuel Annan (Second Author); Yoonbae Jun (Corresponding Author)

A spatio-temporal evaluation of emerging climate risk factors on COVID-19 outcomes in Reno and Las Vegas, Nevada.

Background: Climate change in Nevada has led to multiple health risks, including poor air quality, wildfire, and excess heat stress. Recent studies have provided evidence that both short- and long-term exposure to air pollution and extreme heat may exacerbate COVID-19 outcomes, including increased mortality risk [1-5], but these health effects require further investigation within local communities and subpopulations for better improving public health systems. The project aims to enhance the estimation of the health effects of major climate risk factors, building on the PI’s expertise in data analytics and advanced Bayesian hierarchical methodologies [6-11]. Methods: In this study, we construct a comprehensive dataset encompassing air pollution exposures, heat waves, wildfire smoke, COVID-19 outcomes, and various non-environmental factors with spatial and temporal dimensions. We conduct state-of-the-art statistical analysis to estimate the health effect of the climate risk factors on COVID-19 outcomes in Reno and Las Vegas. We will publish a vulnerability mapping website as well as high-impact journal articles based on the analysis outputs. Implications: Our findings may indicate a significant association or potential causation between long-term exposure to climate risk factors and COVID-19 outcomes in Nevada, highlighting the need for establishing an unified and publicly accessible research framework applicable to various Disease Mapping studies in Nevada.