CDT Module 1: Statistical Computing

Module Outline

Michaelmas Term 2018

We will be using the statistical software package R, which you can get here. Please install it on your own computer and practice using it as early as possible.

We will also use RStudio, which you can obtain here.

The module assumes familiarity with the basic R material found in Oxford's MSc course; some notes can be found here.


R Questions Review Sheet (Solutions)

Air pollution data


Numerical Issues and Computing

Slides on Bayesian Computing, and Adaptive MCMC

Notes on R

Notes on Setting Up an R package with git

Notes on Your Project

  1. Some Points About Functions

  2. Profiling

  3. Speeding Up Code I

  4. Speeding Up Code II

  5. Calling External Code

  6. Debugging

  7. Writing R Packages


Here's an example vignette file and a BibTeX file. It generates a vignette which looks something like this.

Adding Vignettes to Your Package


Hadley Wickham's books on Advanced R and R packages are highly recommended reading.

The R Inferno takes a detailed look at some of R's quirks.

Michaelmas 2014

Talk Slides

Adaptive MCMC Slides (Krys Latuszynski)

Lauritzen and Spiegelhalter (1998) (Robin Evans)

Intractable Likelihoods MCMC (Matti Vihola)

Lasso (Francois Caron)

Scaling in MCMC (Gareth Roberts)

Statistical Challenges in fMRI (Thomas Nichols)