The (Rmarkdown) source code for these lectures is available at the github repository, or by replacing the .slides.html suffix to .Rmd in the link below; the slides are made using reveal.js.

Fall 2019

Week 1 (10/1)

Overview of data science - description and estimation, uncertainty and simulation, with examples for comparing means and linear regression; smoothing.

Week 2 (10/7)

Analysis of Variance (ANOVA) and experimental design; tidy data; power and false positives

  • Reading: Quinn & Keough chapters 5, 7, 8
  • Slides
Week 3 (10/14)

Plotting/visualization, and permutation/bootstrapping

  • Slides
  • Reading: Quinn & Keough chapter 9
Week 4 (10/21)

Multivariate ANOVA, regression, least-squares likelihood

  • Slides
  • Reading: Quinn & Keough chapter 6
Week 5 (10/28)

Model selection; random effects and mixed models - a first look.

  • Slides
  • Reading: Quinn & Keough chapter 13
Week 6 (11/4)

Bayesian statistics - prior distributions and uncertainty - using Stan)

  • Slides
  • Reading: Kruschke, chapters 1, 2, 4, 5, 6, 7
Week 7 (11/11)

Bayesian hierarchical modeling - shrinkage, and sharing power

Week 8 (11/18)

Logistic regression - simulation, posterior predictive sampling - robust regression

  • Slides
  • Reading: Kruschke, chapters 13, 21
Week 9 (11/25)

Robust multiple linear regression; scale mixtures; model selection with crossvalidation; introduction to Generalized Linear Models (GLM).

  • Slides
  • Reading: Kruschke, chapters 15, 16, 17
Week 10 (12/2)

More GLMs; Poisson regression; detecting and modeling overdispersion; quantifying goodness-of-fit.

  • Slides
  • Reading: Kruschke, chapter 24

Winter 2020

Week 11 (1/7)

Survival analysis

Week 12 (1/14)

Time series: temporal autocorrelation, autoregressive models; mechanistic models

Week 13 (1/21)

Categorical data: chi-square for contingency tables, permutation tests.

  • Slides
  • Reading: Kruschke, chapter 16 (metric data with one or two groups), and chapter 24 (Poisson, contingency tables)
Week 14 (1/28)

Categorical prediction; crossvalidation for model comparison.

  • Slides
  • Reading: Kruschke, chapter 22 (softmax regression)
Week 15 (2/4)

Many dimensions: Sparsifying priors and variable selection.

  • Slides
  • Reading: Kruschke, chapters 17 (one-variable linear regression), 18 (multivariate linear regression)
Week 16 (2/11)

Many response variables; data analysis example (cream cheese)

Week 17 (2/18)

Factor analysis, dimensionality reduction, and visualization; clustering; PCA, PCoA, MDS, t-SNE, UMAP

Week 18 (2/25)

Latent factors, deconvolution for mixtures of expression data; nonnegative matrix factorization

Week 19 (3/3)

Deconvolution continued; introduction to spatial statistics and kernel density estimation.

Week 20 (3/10)

Spatial statistics: spatial autocorrelation and Kriging.