Bayesian Biostatistics, 25-27 Jan 2016, iDiv, Leipzig, Germany

The course is full.

• Syllabus and instructions.
• Complete raw codes (Markdown and R) and materials see the course's GitHub repository.

DAY 1

1. Introduction: Course contents, pros and cons of Bayes, necessary skills.
2. Normal distribution: Introducing likelihood and deviance on the Normal example.
3. Poisson distribution: The didactic simplicity of Poisson and its likelihood. Likelihood maximization. Probability mass function.
4. The Bayesian way and the principle of MCMC sampling.
5. Simplest model in JAGS. Estimating lambda parameter of simple count data using MCMC.
6. Bayesian resources: Overview of software, books and on-line resources.

DAY 2

1. T-test: First model with 'effects', hypotheses testing, derived variables.
2. Linear regression part 1 and part 2: Extracting credible and prediction intervals.
3. ANOVA part 1 and part 2: Fixed vs. random effects, the effect of small sample, shrinkage.

DAY 3

1. Simple autoregressive (AR) and generalized additive models (GAM), time series analysis.
2. Site occupancy model: logistic regression, imperfect observation, latent variables.
3. Useful probability distributions (binomial, beta, gamma, multivariate normal, negative binomial).
4. Publishing Bayesian papers.
5. Concluding remarks.

RELATED TOPICS AND POSTS

• Survival analysis.
• Spatial model using GAM splines in JAGS.
• Previous year's lesson on time series analysis using the lynx dataset.
• Poisson regression using GLM, maximum likelihood and JAGS.
• Tailoring custom distributions.
• Bayesian Principal Components Analysis (bPCA)