My research interests are diverse. One line of research focuses on methods for spatio-temporal data -- graphical models, spatial machine learning (Gaussian Processes and random forests), fast Bayesian algorithms for high-dimensional spatial data. These methods research usually emerge from my collaborations with researchers in areas of air pollution and environmental health. One exciting current project involves statistical calibration of hyper local air-pollution data from low-cost monitors.
I also work on hierarchical models, Bayesian machine learning and shrinkage (regularization) methods for complex survey-based datasets arising from epidemiological field work. Current applications include small area estimation using misaligned and partly-missing survey summaries, quantification learning of child mortality rates.
If you're interested in any of the areas and would like to discuss research or collaboration opportunities, feel free to email me or drop by my office.