Highlights

Authored 48 manuscripts

36 in peer-reviewed scientific journals
12 internal technical reports

34 Presentations

24 at scientific meetings
10 technical presentations

Statistical consulting & technical assistance

on 100+ scientific projects as a non-author

Developed software packages

using R, Python, C, SQL, SAS, C#, Visual BASIC

Subject matter expert

in statistics & data mining for internal/external workgroups and mid-level leadership

Invited Reviewer

for 6 scientific journals

Education

University of Georgia2001, 2004, 2007, 2007


PhD Forest Biometrics, 2007

Awarded the Jerome L. Clutter Memorial FELLOWSHIP for outstanding student research

Inducted into the Xi Sigma Pi International Forestry HONOR SOCIETY

Dissertation: Sampling efficiencies of new and existing variable radius plot growth estimators

MS Statistics, 2007

Coursework including Probability Theory I & II, Mathematical Statistics I & II, Non-linear Regression, Time Series Analysis, Numerical Computing

MS Forest Biometrics, 2004

Thesis: Using a camera as an angle gauge in angle-count sampling

BS Forestry, 2001

With an emphasis on quantitative decision making

Selected Projects

Bayesian Model Averaging

PDF slides

I used Bayesian Model Averaging in R for model selection, to assess variable importance, predict new cases, and assess model uncertainty in multiple scientific projects.




Sample Size Web App

GitHub Repo

I used Python and R to create a web application that aids researchers in the deign of multi-stage, stratified cluster surveys. It computes the needed sample size -- or tells the researcher what size confidence interval can be achieved for a given sample size -- and outputs tables, graphs, and summary statements suitable for use in protocols and manuscripts.




A/B Testing Algorithms

PDF poster

I invented algorithms to quickly compute the Type I error rate of sequential A/B tests under both a Binomial & Poisson design. I coded the algorithms in C, with optional interfaces in R or as a stand-alone DLL. My code also computes exact-test critical value, statistical power, & lists decision-making data points. Runtime < 1 second -- 3 seconds; the code it replaced required overnight-to-days.




Ethereal to Make Docs

GitHub

A framework using Python to convert a collection of Markdown files into a website.

Selected Manuscripts

Continuous sequential boundaries for vaccine safety surveillance

Winner of CDC's Best Theoretical Paper award

Time-varying boundaries to controll type I error over the course of sequential A/B testing

Statistics in Medicine 33(19)

A Bayesian Approach to Sequential Analysis in Post-licensure Vaccine Safety Surveillance

Winner of CDC's Best Theoretical Paper award

The frequentist FPR was better, but the Bayesian time-to-signal & FNR was lower

Pharm Stat. 19(3)

PDF

A Bayesian approach for modelling non-linear longitudinal/hierarchical data with random effects in forestry

Highlighted the unified manner in which estimation and prediction are handled

Forestry 85(1)

Comparing Traditional Longitudinal Modeling Strategies of Forestry with Mixed-Effects Models: Restrictions in Model Formulation

The bias-variance tradeoff is better handled in predictive modeling with guarded parameter-space flexibility

Forest Science 57(5)

PDF

Selected Presentations

Interfacing R and C

PDF slides

The best of both worlds. Sometimes it makes sense to write your own C modules. In this talk I shared with colleagues how we can use our C code in R scripts.




An Introduction to Bayesian Statistics

PDF slides

Bayesian statistics represent a powerful paradigm for all forms of statistical inference. I gave this introductory-level talk to fellow scientists.




Reproducible Research Tools

PDF slides

The days before Jupyter Notebooks. This is a talk I shared with colleagues on existing methods to do reproducible research.

Awards and Accolades

  • Two-time winner of BEST THEORETICAL PAPER, CDC’s Statistical Science Award for Statistical Excellence: Continuous Sequential Boundaries for Vaccine Safety Surveillance Stat in Med (33):19
    A Bayesian Approach to Sequential Analysis in Post-licensure Vaccine Safety Surveillance Pharm Stat (19):3

  • HONOR AWARD, National Center for Emerging and Zoonotic Infectious Diseases (NCEZID), for:
    Excellence in rapidly assessing the safety of Tdap vaccination in pregnant women.

  • CERTIFICATE OF APPRECIATION, CDC Division of Tuberculosis Elimination, for:
    Exemplary collaboration & innovation in quantitative science to respond to the leveling of Tuberculosis incidence.

  • DIRECTOR’S SPECIAL ACT IN RECOGNITION, NCEZID/Division of Healthcare Quality Promotion, for:
    Development of innovative methods to aid Rapid Cycle Analysis vaccine safety surveillance activities.

Terms of Use

The purpose of this website is to serve as personal branding and professional networking for the website owner. The website owner retains all rights to the original content, design, and code, and users are not permitted to copy, reproduce, or distribute the site's content without explicit permission. Visitors are expected to use the site responsibly and ethically. The opinions expressed here are those of the website's owner; mention of a product or organization does not imply endorsement. The information provided on this website is for informational purposes only and the website owner is not liable for any errors or omissions in the content. The website owner reserves the right to modify or remove content at any time without notice.

Learn More

Linkedin
GitHub
Google Scholar
Printable Resume