- 1. lecture: Modelling and latent variables.PDF version.
- 2. lecture: Inference on latent variable models.PDF version.
- 3. lecture: First Bayesian lecture. PDF version.
- 4. lecture: Sampling using MCMC. PDF version.
- 5. lecture: Second Bayesian lecture. PDF version.

- Exercise description
- salamanders_ml.R - R support code for exercise 2 and 3.
- salamanders_binom.wb - WinBUGS code for binomial treatment of the salamander data. Can be used as a template for exercise 4.
- salamanders.wb - WinBUGS code for occupancy treatment of the salamander data. Can be used as a template for exercise 4.
- salamanders_testbinom.wb - WinBUGS code for model testing (binomial vs occupancy model).
- weta_binom.wb - Binomial treatment of Weta data.
- weta.wb - Occupancy treatment of Weta data.
- weta_testbinom.wb - Testing binomial vs occupancy mode for Weta data.
- weta_testbrowsed.wb - Testing occupancy with vs without the explanationv ariable 'browsed' (goat browsing) for Weta data.
- weta_testbrowsed_big.wb - Testing occupancy and binomial model with vs without the explanationv ariable 'browsed' on the various parameters (6 models in total) for Weta data. Seems to only work for WinBUGS, not OpenBUGS.

- normal_model.wb - Normal model BUGS code.
- mean.wb - Normal model BUGS code without data.
- normal_modelchoice.wb - Testing mu=560 for normal model.
- regression.wb - Linear regression in BUGS.
- rating.wb - Non-linear regression example using BUGS. The example is for water flow vs water level in a river cross-section, with log(flow)=a+b*log(water level-river bottom) (meaning a power-law for the flow, but with unknown zero-level.) This example was used in one of the figures in order to see the effect of high dependency between parameters.
- testbinom.wb - Tests binomial vs occupancy for an artifical dataset that was made using the binomial distribution.

- weta_winbugs.R - Example code for using BUGS from R.
- dragons.R - R code for calculating the probabilities in exercise 1.
- weta_ml.R - ML code for the Weta data.
- trees.R - Code for the trees example.
- rw_boundy.R - Creates a Markov Chain like the one described in the MCMC lecture. (Demonstrates that a normal random walk with reflective bounry gives a uniform distribution).

- Back to my main page.