Date(s) - 05/13/2019
3:30 pm - 5:00 pm


3910 15th Ave NE
Seattle WA

eScience affiliate Adrian Dobra will give a talk titled “Human mobility: Bayesian and frequentist insights from big data.”


In this talk I will focus on the statistical modeling of human mobility. I will introduce human activity spaces which are fundamental to the assessment of individuals’ dynamic exposure in demography and public health. I will survey existing approaches for measuring the geometry, size and structure of human activity spaces based on GPS data, and explain their limitations. I will discuss model selection and inference with graphical loglinear models in hyper-sparse contingency tables. I will show the role of priors that automatically handle the massive multiple comparisons problem related to the selection of large graphs. I will illustrate the practical value of these methods for modeling human mobility using geolocated Twitter data and with a recently collected GPS dataset that comprises the locations visited by ten individuals over a six months period.

More information on this seminar series available here:

This event is open to the public.