During this presentation, Joe will outline why a researcher might choose to analyze their complex data sets with Bayesian inference, over frequentist (i.e., more conventional) approaches. Joe will show a brief example of conducting a multiple linear regression and ranking predictor influence, using the statistical programming language Stan, which he will interface with R packages `rstan` and `rstanarm`.
Joe is an Assistant Professor in SICCS, and his research focuses on building and testing mathematical models that represent how environmental heterogeneities influence the spread of infectious diseases in wildlife and human populations. For example, one NSF-funded project seeks to understand the interactions between climate change and infectious disease in explaining long-term stability of ectotherm populations. One NIH-funded project is building software to implement and test spatial models of infectious disease transmission of humans, to increase the standardization and fidelity of model forecasts. Joe received his BA in Environmental Sciences from WashU in St. Louis, got his PhD in Ecology & Evolutionary Bio at Univ of Colorado-Boulder, and then did a postdoc at Univ of Chicago, and one at Univ of California-Santa Barbara before coming to NAU in 2018.