Stats 253: Analysis of Spatial and Temporal Data

The following are archived materials from Summer 2014.

Theme	Monday	Wednesday
Introduction and autoregression	What is spatial and temporal data? Review of linear regression Autoregressive processes Lecture Slides Reading: Sherman, Ch. 1, Shumway/Stoffer, Chs. 1, 2	Bootstrap standard errors for AR processes Spatial autoregression Lecture Slides Reading: Sherman, Ch. 4 J. Besag (1974), Spatial Interaction and the Statistical Analysis of Lattice Systems (with discussion).
Visualizing data, state-space models	R Tutorial Intro to R Spatial statistics packages Lecture Materials Reading: skim Bivand, Chs. 2, 3, 9 (optional: Ch. 4)	state-space models multivariate normal distribution Kalman filter Lecture Slides Reading: Shumway/Stoffer, Ch. 6.1-6.3 G. Petris and S. Petrone (2011), State Space Models in R
Geostatistics	more on the Kalman filter MMSE prediction geostatistics / kriging Lecture Notes Reading: Sherman, Ch. 2	stationarity covariance estimation variogram Lecture Slides Reading: Sherman, Ch. 3; Bivand, Ch. 8
Frequency domain methods	Fourier transform Spectral analysis Lecture Slides \| R Code for sinusoids, noise Reading: Shumway/Stoffer, Ch. 4.1-4.6	Time-frequency analysis Non-negative matrix factorization Audio source separation Lecture Slides \| Audio: Mary, Madonna R functions, R scripts: Mary, Madonna Reading: Smaragdis et al (2014), Static and Dynamic Source Separation Using Nonnegative Factorizations.
Point processes	Poisson processes Second-order interactions Lecture Slides R Code for CSR demo Reading: Sherman, Ch. 7; Bivand, Ch. 7	Spatio-temporal point processes Lecture Slides Reading: Gabriel et al (2012), stpp: An R Package for Plotting, Simulating and Analyzing Spatio-Temporal Point Patterns
Spatio-temporal Processes	Testing for isotropy Kriging for spatio-temporal data Connection with kernel methods Lecture Slides Reading: Sherman, Ch. 6	Dynamic Autoregressive Spatio-Temporal Models Lecture Slides Code: Wind Data Analysis
Project Presentations (Thursday and Friday)	No Lecture (Joint Statistical Meetings)	No Lecture (Joint Statistical Meetings)
Bayesian Approaches and Hierarchical Models	Fundamentals of Bayesian inference Hierarchical models Introduction to BUGS / STAN Lecture Slides Peer Grades Example: (Shuffled) Data \| JAGS \| R Code	Hierarchical models for spatio-temporal data Lecture Slides Reading: