Lecture 1 (Wednesday, August 20): Statistical experiments versus observational studies; the analysis of statistical experiments—analysis of variance
Lecture 2 (Lab 1) (Monday, August 25): Three-factor analysis of variance
Lecture 3 (Wednesday, August 27): Probability distributions useful in ecology I: normal, lognormal, gamma, beta, Bernoulli
Lecture 4 (Wednesday, September 3): Probability distributions useful in ecology II: binomial, multinomial, Poisson, negative binomial
Lecture 5 (Lab 2) (Monday, September 8): Three-factor analysis of variance (continued)
Lecture 6 (Wednesday, September 10): Maximum likelihood estimation
Lecture 7 (Lab 3) (Monday, September 15): Maximum likelihood estimation
Lecture 8 (Wednesday, September 17): Bayesian estimation
Lecture 9 (Lab 4) (Monday, September 22): Bayesian estimation using WinBUGS
Lecture 10 (Wednesday, September 24): Information-theoretic tools for model selection
Lecture 11 (Lab 5) (Monday, September 29): Using AIC for model selection
Lecture 12 (Wednesday, October 1): Goodness of fit tests. Introduction to generalized linear models
Lecture 13 (Lab 6) (Monday, October 6): Goodness of fit tests for count data models
Lecture 14 (Wednesday, October 8): Predictive simulation; Generalized linear models II.
Lecture 15 (Lab 7) (Monday, October 13): Predictive simulation: frequentist and Bayesian versions
Lecture 16 (Wednesday, October 15): Generalized linear models III; logistic regression
Lecture 17 (Lab 8) (Monday, October 20): Fitting logistic regression models in R; goodness of fit tests
Lecture 18 (Wednesday, October 22): Goodness of fit for logistic regression; sensitivity, specificity and ROC curves
Lecture 19 (Lab 9) (Monday, October 27): Multiple logistic regression; sensitivity, specificity, ROC curves, cross-validation
Lecture 20 (Wednesday, October 29): Analysis of correlated data: random effects models
Lecture 21 (Lab 10) (Monday, November 3): Hierarchical random effects models in R
Lecture 22 (Wednesday, November 5): Strategies for model building with multilevel models
Lecture 23 (Lab 11) (Monday, November 10): Hierarchical random effects models in WinBUGS
Lecture 24 (Wednesday, November 12): Bayesian estimation of random effects models, random versus fixed effects, how random effects account for heterogeneity and correlation
Lecture 25 (Lab 12) (Monday, November 17): Lattice graphics for multilevel models. Adding an additional residual correlation structure to random effects models.
Lecture 26 (Wednesday, November 19): Methods of estimation in random effects models: full likelihood, restricted maximum likelihood, and the Bayesian alternative. Hierarchical models with more than two levels: R syntax. Mixed effects nonlinear models.
Lecture 27 (Lab 13) (Monday, November 24): Deciding which level-1 parameters should be made random. Fitting mixed effects nonlinear models with the nlme package of R.
Lecture 28 (Lab 14) (Monday, December 1): Fitting mixed effects nonlinear models with the nlme package of R. Using lattice to display the fitted model.
Lecture 29 (Wednesday, December 3): Accounting for variance heterogeneity in fixed effects and mixed effects models with a positive response. Obtaining the log-likelihood of mixed effects models with a transformed response.
Course Home Page
Phone: (919) 962-5930
Address: Curriculum in Ecology, Box 3275, University of North Carolina, Chapel Hill, 27516
Copyright © 2008
Last Revised--September 8, 2009