Teaching

In Fall 2019, I developed and taught a new course in Dimension Reduction for the Department of Statistics at Harvard University intended for upper level undergraduates.  Since then, the focus has shifted to more general methods for unsupervised learning.

Course notes on select topics will be made available throughout the fall.  A overview of the course, presented to students during the first meeting, is available here.

• • Images contained in these introductory slides are supplied with relevant references to source content.

  • Special thanks to Paul Bendich, Duke University, for his insight on content and course goals.

  • For full details see the course's GitHub site with lessons and numerical demos of selected techniques from the first iteration of the course.

I developed an introductory course in Spatial Statistics, which was first offered during the Spring 2021 semester (Harvard's second fully remote semester).  We cover point processes, areal data, and geostatistics balancing theory (and assumptions), techniques for exploratory analysis, and inference and prediction.  Numerical tutorials, provided to the class, are provided to the course.  Techniques for Bayesian methods with MCMC are covered at the end of the course.

• Point processes 

  • Uses sf for visualization; spatstat for EDA and inference

• Areal data

  • Uses sp, sfdep for spatial weights and EDA; spautolm for regression

• Geostatistics

  • Uses sf for visualization, gstats for EDA and kriging

Data sources include spData and NYCOpenData.