Daniel was Statistics; Environmental Science, Policy, & Management postdoctoral scholar and a BIDS Data Science Fellow. Daniel’s background and interests lie at the intersection of mathematics and computer science. He completed his PhD in Statistics at the University of Otago, New Zealand, in the area of model uncertainty and model averaging. His dissertation proposed and studied a new methodology for constructing confidence intervals for parameters common to a set of candidate models in the presence of model uncertainty. Daniel is currently an Assistant Professor of Statistics at Williams College.
After completing his PhD, Daniel joined an interdisciplinary team at Berkeley developing a flexible algorithmic package for use in R. This software, NIMBLE, takes a new approach toward the statistical analysis of hierarchical models. It provides a framework for specifying statistical algorithms that may subsequently be applied to any model and dataset. This approach facilitates the application of a wide range of statistical algorithms to any particular model while also facilitating the customization of existing statistical algorithms.
Daniel’s research interests are in computational statistics, hierarchical model analysis, and MCMC sampling algorithms. This includes studying existing sampling algorithms; their relative efficiencies in relationship to hierarchical model structure; and potential modifications to existing sampling algorithms, including hybrid designs.
Daniel’s applied work has generally been in the area of statistical ecology, including such topics as population dynamics, occupancy modeling, and capture-recapture analyses.