Spatial and Spatial Temporal Statistics: Modeling and Applications in R

Instructor: Aritra Halder, PhD, MS, Assistant Professor of Biostatistics, Dornsife School of Public Health

Dates: Monday, June 23 - Friday, June 27

Times: 9:00 a.m. - 1:00 p.m. EST

Format: Hybrid instruction

A comprehensive introduction to the statistical methods used in the analysis of geo-referenced spatial data. The course covers the topics of disease mapping (relative risk estimation), disease clustering, ecological analysis. The methods covered are mainly in the area of generalized linear models and mixed models. The course addresses the use of appropriate software packages for the analysis of disease incidence data. The progression of methods begins with simple Poisson regression(log-linear models) and logistic linear models and moves to Bayesian hierarchical modeling for mapped data and finally to models with spatially correlated prior distributions only available in advanced software. If time permits, we also examine space-time modeling, multivariate analysis and survival modeling. Knowledge of intermediate statistics and basic proficiency in R is expected.

Learning Objectives:

1. Analyze the variety of data found in spatial epidemiological studies

2. Apply the R software packages to spatial epidemiological analyses

3. Demonstrate an understanding of the theory underlying the appropriate concepts and methods

Continuing Education Credits*: 1.5 CEU or 15 CPH