June 12, 2025
Remembering the Double Poisson Distribution
Why you might want to consider using the double Poisson in place of the Conway-Maxwell Poisson (CMP) Distribution.
Read MoreAbout Me
Hi! I'm Nate Byford, currently pursuing a Ph.D. in Statistical Sciences at Baylor University, with a concentration in Data Science. Over the past five years, I've leveraged advanced statistical methods and machine learning to transform complex data into actionable insights - completing numerous consulting projects, helping to author papers and conference presentations, and consistently delivering client satisfaction.
My passion lies at the intersection of statistical analysis, data-driven decision-making, and real-world impact. Whether it's designing Bayesian A/B tests, developing predictive analytics, or implementing cutting-edge AI/ML, I strive to make the world a safer, smarter place through data science.
Technical Expertise
Statistical Analysis
Machine Learning
Programming
Visualization
Data Engineering
Research Methods
Featured Projects
Detecting Benford's Law
Developed and tested innovative methods for detecting Benford’s Law in real data, advancing statistical analysis techniques.
R
Simulation Studies
NLP
Correcting under-reporting in over-dispersed spatial count data
Developed and implemented Bayesian method to model under-reported and over-dispersed spatial counts in R with nimble.
R
Spatial Statistics
Bayesian Methods
Latest Blog Posts
February 21, 2023
Setting up Linux for Statistics and Data Science
Quick start for how to setup Linux for data science and statistical work.
Read MorePublications
Assessing the Effects of Surface-Stabilized Zero-Valent Iron Nanoparticles on Diverse Bacteria Species Using Complementary Statistical Models
This study investigated how surface coatings affect the antimicrobial properties of zero-valent iron nanoparticles (FeNPs). Using a bottom-up synthesis approach, FeNPs were coated with L-ascorbic acid (AA), cetyltrimethylammonium bromide (CTAB), or polyvinylpyrrolidone (PVP), then tested against six diverse bacterial species. Disc diffusion assays showed AA- and CTAB-coated FeNPs had the strongest antibacterial activity. Sensitivity varied by species, with Bacillus nealsonii being the most susceptible. Statistical models confirmed significant differences in bacterial response across coatings and concentrations, reinforcing the potential of customizable 'designer nanoparticles' for targeted antimicrobial use.