WSU Math/Stats Department News & Events: 2023

Using Statistical Learning to Predict Bald Eagle Flight from Land Features to Inform Safe Wind Turbine Placement

Silas Bergen, PhD

Data Scientist and Biostatistician
Assistant Professor in Math/Stat Dept.

Abstract: As wind energy development increases, so does the potential for bird fatalities from collision with turbine rotors. In this talk I will describe how I used statistical learning methods to analyze millions of data points collected from GPS telemetry devices attached to bald eagles. The intent of this research was to understand land types where bald eagles might be at greater risk of collision with industrial wind turbines to inform safe placement of wind farms.

Friday, December 1st

12:15- 12:40 PM

SLC 120

A Brief Look at Statistical Genetics from a National Cancer Institute Postdoc

Dr. Jacob Williams

National Cancer Institute

Abstract: Statistical genetics is a wide-ranging field that at its core is attempting to quantify the relationship between variations in DNA and observable traits/diseases. As the majority of DNA is similar across individuals of common ancestry, the differences in DNA are modeled using regression techniques. However, a complex part of modeling these differences is the that the number of differences is quite large. For example, there are an estimated 4 to 5 million SNPs in the human genome. Further, some of these differences occur at an extremely low rate so to investigate these rare differences sample sizes of up to 500k individuals are obtained. With this p>>n problem and datasets in the petabytes, techniques to perform causal inference and risk prediction must be specially designed. As a doctoral student I developed methods to accurately perform causal inference and as a postdoctoral fellow I am currently developing methods to increase risk prediction accuracy using both common and rare differences. In this talk, I will provide a high-level overview of these methods as well as a background of how I became a statistical geneticist and a high-level overview of statistical genetics.

Wednesday, November 15^th

12:00- 12:50 PM

Gildemeister 155