An introduction
I am an R&D mathematical statistician working on scientific problems related to all things forest management. The data is often multivariate and longitudinal in space/time, and expensive to collect (i.e., time, labor, resources). It comes from many kinds of studies—designed experiments, observational surveys, and surveys of convenience and opportunity. As such, my work requires broad statistical expertise to meet the needs of many projects and partners.
The work I do
Most projects involves bread-and-butter statistical and machine learning techniques like generalized linear mixed models, decision trees, and random forests. These are powerful tools that are well understood and perform well when data is small or of modest size.
For the most complex projects I use Bayesian variants of models found in statistical physics (e.g., the Ising model). Their interpretation can be done in terms of simple physical systems like circuits and magnets.
For more information, see my portfolio.
My training
Before working in forest management R&D, I received my PhD in statistics from Virginia Tech and had postdoctoral fellowships in statistical ecology at Colorado State University and NIMBioS (now NIMBioS). As a graduate student, my assistantship was funded by working as a lead statistical collaborator at LISA (now SAIG). I also received an NSF Math-Bio Undergraduate Fellowship as an undergraduate mathematics student at UNCG.
For more information, see my CV.