April 25, 2016 | Erin Bluvas, email@example.com
Associate Professor Marco Geraci made his first appearance at the Arnold School when he was a doctoral student studying Applied Statistics at the University of Florence in his native Italy. At the time, he was a visiting scholar who was beginning to develop his ideas for an original statistical model that would eventually be applied by researchers from across the world.
In 2005-2006, he returned to the Arnold School as a postdoctoral fellow and continued to build his ideas for what he later named Linear Quantile Mixed Models—publishing his first peer-reviewed paper on the topic in 2007. Since then, Geraci’s models have been applied by researchers and institutions from a range of disciplines (e.g., medicine, public health, ecology, finance, linguistic and lexicography) around the globe, including the National Aeronautics and Space Administration (NASA), and numerous universities and other research institutions (see a sample of these studies).
“One of the key factors that helped spread Linear Quantile Mixed Models is the availability of related computer software, freely available under the GNU General Public License” says Geraci. “Because of this access, the software has been downloaded almost 10,000 times since 2012.”
After spending the past ten years conducting research at academic institutions in the United Kingdom and publishing about his models, Geraci’s path led him back to the Arnold School where he is an associate professor of biostatistics in the Department of Epidemiology and Biostatistics. A recently awarded $146k R03 grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development will enable Geraci to take his models even further. His next step will be to extend the Linear Quantile Mixed Models to nonlinear cases to develop Nonlinear Quantile Mixed Models.
“This project aims to develop advanced statistical methods to address some complex issues we encounter is various fields of research,” says Geraci. “There will obviously be some modeling and computational challenges associated with this endeavor. Quantile regression modeling has gained momentum in many applied fields, including public health, and this a good time to push further the applicability of these methods.”
Another important aspect of this project is that it will use data from pediatric research to test the new methods. “I’ve been working with experts in child health since 2007,” says Geraci, who collaborates with child health epidemiologists at USC and is a member of the Research Consortium on Children and Families. “So this is a natural extension of some of my other research projects and will provide some very valuable insights related to pediatric outcomes and treatment.”
Geraci’s proposed research will advance models that are currently applied in public health to assess associations between risk factors (e.g., blood lead levels) and health outcomes (e.g., blood pressure). “There is a tremendous need to model the data appropriately in order to more accurately quantify medical and economic costs associated with a given risk factor, or to assess the benefits of a medical treatment for patients who deviate from the average,” Geraci explains. “The key element of the proposed models is to provide a framework of analysis for assessing nonlinear effects of exposures on health outcomes in subjects who rank lower or higher in the distribution—for example, those who are preterm and low birthweight, obese, sedentary, hypotensive, hypertensive, or socio-economically disadvantaged. These are individuals that may experience higher mortality, morbidity or social exclusion and may require targeted interventions.”
Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute Of Child Health & Human Development of the National Institutes of Health under Award Number R03HD084807. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.