OZ Statistical Consulting | Spatial & Spatio-Temporal Analysis Experts

What We Do

Statistical Consulting Services

From exploratory analysis to advanced modeling, we provide comprehensive statistical support for your research and business needs.

Spatial Analysis

Geostatistics, kriging, spatial autocorrelation, point pattern analysis, and geographic data modeling.

Spatio-Temporal Modeling

Dynamic models that capture both spatial dependencies and temporal evolution in your data.

General Statistics

Regression, hypothesis testing, experimental design, Bayesian analysis, and machine learning.

Research Support

Study design, power analysis, manuscript preparation, and peer review response support.

Specialization

Deep Expertise in Spatial Statistics

Spatial and spatio-temporal data present unique challenges that require specialized methods. Our core expertise lies in understanding and modeling the complex dependencies inherent in geographic data.

Gaussian Processes Flexible non-parametric models for spatial interpolation and prediction
Point Process Models Analysis of spatial point patterns and event occurrences
Hierarchical Spatial Models Bayesian frameworks for complex spatial structures
Space-Time Dynamics Models capturing evolution of spatial patterns over time

spatial_gp.R

# Spatial Gaussian Process
# Squared Exponential Kernel

kernel <- function(x1, x2, l, sigma) {
  d <- sqrt(sum((x1 - x2)^2))
  sigma^2 * exp(-d^2 / (2 * l^2))
}

# Build covariance matrix
K <- outer(locations, locations,
           Vectorize(kernel),
           l = 0.5, sigma = 1.0)

# Sample from GP prior
f <- mvrnorm(n = 1,
             mu = rep(0, n),
             Sigma = K)

About

Dr. Aaron Osgood-Zimmerman

Spatial statistician, researcher, and educator with over a decade of experience turning complex geographic data into actionable insights.

I'm an Assistant Professor of Statistics at Bucknell University and the founder of OZ Statistical Consulting. My research focuses on building bespoke spatial statistical modeling frameworks tailored to specific applications and real-world inferential questions.

Before joining Bucknell, I spent four years at the Institute for Health Metrics and Evaluation (IHME) at the University of Washington, where I led a team of statisticians and software engineers developing high-resolution spatio-temporal models. Our work produced 5×5km resolution maps of child health outcomes across Africa and low- and middle-income countries, published in Nature, Nature Medicine, The Lancet, and the New England Journal of Medicine. This work was cited by Kofi Annan in a Nature World View letter.

I hold a Ph.D. and M.S. in Statistics from the University of Washington, where I worked with Jon Wakefield, and a B.A. in Mathematics and B.S. in Engineering from Swarthmore College.

My current research extends spatial statistical methods to spatial transcriptomics and cancer epidemiology, and I continue to develop novel approaches to Gaussian process modeling and Bayesian spatial analysis.

Education

Ph.D., StatisticsUniversity of Washington, 2022
M.S., StatisticsUniversity of Washington, 2015
B.A., Mathematics & B.S., EngineeringSwarthmore College, 2011

Selected Publications

Lead author, Nature (2018)
Co-lead author, Nature (2020)
Co-lead author, Nature Medicine (2021)
Lead author, Intl. Statistical Review (2023)

Links

Google Scholar ORCID Download CV (PDF)

Transform Your Spatial Data Into Actionable Insights