A short list of the resources most useful for this course. Please let me know if there are other links that you think should be included on this page!
Weekly "Stats Chats" at the UNR library: drop-in help with statistics and R.
Math and probability background
Short, visual introductions to the math this course uses. Good places to start if a topic feels shaky.
- Essence of Calculus (3Blue1Brown): derivatives, integrals, and the fundamental theorem, with almost no algebra.
- Essence of Linear Algebra (3Blue1Brown): vectors, matrix multiplication, determinants, inverses, and eigenvalues as geometry.
- Probability (3Blue1Brown): includes Bayes' theorem, the central limit theorem, and why the normal distribution has a \(\pi\) in it (the Gaussian integral).
- Seeing Theory: interactive introduction to probability, distributions, inference, and regression.
- The Distribution Explorer: the common distributions, what stories generate them, and how they relate, with interactive plots.
Mathematical statistics
- The geometry of statistics (Sam Levey): how linear algebra explains regression, variance, and more; a good follow-up to the linear algebra videos.
- STAT 414: Introduction to Probability Theory and STAT 415: Introduction to Mathematical Statistics (Penn State): free course notes covering random variables, expectation, distributions, likelihood, and maximum likelihood estimation, with worked examples.
- Statistical Rethinking (McElreath): book and free lecture videos. A readable, ecology-friendly path through likelihood, Bayesian inference, multilevel models, and causal reasoning.
Bayesian inference, MCMC, and Stan
- Bayes Rules! (Johnson, Ott, and Dogucu): a free introductory Bayesian textbook with R code.
- Interactive gallery of MCMC samplers (Chi Feng): watch Metropolis, Gibbs, HMC, and NUTS explore different posteriors.
- MCMC sampling for dummies (Wiecki): a Metropolis sampler built from scratch.
- A conceptual introduction to Hamiltonian Monte Carlo (Betancourt): why HMC works, with very little math.
- Stan: the documentation (user's guide, functions reference) and the getting started with cmdstanr vignette.
- brms: fit Stan models with
lme4-style formulas.
- Bayesian workflow (Gelman et al.): building, checking, and comparing models in practice.
- NIMBLE and JAGS: BUGS-language samplers, useful for models with discrete latent variables.
Causal inference
- DAGitty: draw a causal diagram and find which variables to adjust for.
- The Effect (Huntington-Klein): a free, readable book on causal diagrams and research design.
Time series and spatial data
Data sources
Public data sets for final projects:
- Dryad: data behind published papers.
- EDI Data Portal: LTER and other long-term ecological data.
- DataONE: searches many environmental data repositories at once.
- NEON data portal: standardized ecological data from sites across the US.
- GBIF: species occurrence records worldwide.