REQUIRED COURSE LEARNING OUTCOMES:

 1. Distinguish different types of spatial data (geostatistical, aerial, point process) to evaluate how spatial autocorrelation plays a role in statistical modeling.

 2. Explore existing methods to investigate spatial autocorrelation in example datasets provided as exercises.

 3. Categorize spatial methods used in research and implement them using statistical software and GIS.

 4. Critically evaluate, defend and apply spatial analytical methods.

 5. Design solutions for local, neighborhood, and regional analyses problems

 6. Generate linearly referenced features and incorporate them into GIS analysis

 7. Evaluate concepts of geostatistical models to interpolate 3-dimensional data

 8. Solve network problems through network analysis

 9. Model geographic distributions

 10. Identify and interpret spatial patterns and clusters

 11. Analyze spatial relationships

 12. Evaluate and interpret spatial statistical results

 REQUIRED TOPICAL OUTLINE

 I. Background and Motivation

         a. Types of data

         b. Preliminary concepts

         c. Spatial structures and modeling

 II. Data Types and Applications

         a. Point level models

         b. Spatial point processes.

         c. Areal (lattice) models

 III. Estimation and Modeling of Spatial Correlations

         a. Estimating variogram

         b. Fitting parametric models

         c. Maximum likelihood estimation

         d. Restricted maximum likelihood

 IV. Prediction and Kriging

         a. Lagrange multiplier approach

         b. Conditional inference approach

         c. Predicting at multiple sites

         d. Model misspecification in kriging

 V. Spatial-Temporal Models

         a. Separable vs nonseparable models

         b. Continuous time models when spatial dependence is nuisance

         c. Spatial models when time dependence is nuisance

         d. Misalignment

         e. Data integration