STA 521 Predictive Modelling and Statistical Learning (Fall 2018)

Course Texts

• An Introduction to Statistical Learning: with Applications in R Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani Get it @Duke
• Data Analysis using Regression and Multilevel/Hierarchical Models. Andrew Gelman and Jennifer Hill Amazon

• Hoff, P. L. (2009), A First Course in Bayesian Statistical Methods, Springer. ISBN 978-0-387-92299-7 get it @ Duke

• Weisberg, S. (2014) Computing Primer for Applied Linear Regression Using R (Useful for R)

• Weisberg, S. (2013) Applied Linear Regression get it @ Duke

Linear/Matrix Algebra

• Gilbert Strang’s Online Course at MIT
  • Video Lectures
  • Introduction to Linear Algebra. Strang, Gilbert. 4th ed. Wellesley, MA: Wellesley-Cambridge Press, 2009. ISBN: 9780980232714. Buy @ Amazon
• Matrix Algebra from a Statistician’s Perspective. Harville, David A. eBook in Duke Library

---

R Programming Language

R is a statistical programming language that is especially powerful for data exploration, visualization, and statistical analysis. To interact with R (initially), we will primarily be using RStudio, an interactive development environment (IDE).

We will mostly be using a browser based version of RStudio on a remote server but you can also install a local version of the RStudio IDE.

Books & Resources for learning R

• Codeschool - Try R A brief R tutorial, in case you would like to have another avenue by which to get introduced to R.
• R for Data Science - Grolemund, Wickham O’Reilly, 1st edition, 2016 (ISBN:)
• ggplot2: Elegant Graphics for Data Analysis Use R! Buy it at Amazon
• Advanced R - H. Wickham Chapman and Hall/CRC, 1st edition, 2014 (ISBN: 978-1466586963)
• R Packages - H Wickham O’Reilly, 1st edition, 2015 (ISBN: 978-1491910597) *The R Project for Statistical Computing
• R Downloads Duke Mirror with Linux, Mac, Windows
• Rstudio Easy user interface for R/R Markdown and more for Linux, Mac and Windows Download
• An Introduction to R (pdf) (html version) the most up-to-date official R intro
• twotorials: how to do stuff in r. two minutes or less.

---

CRAN Comprehensive R Archive Network

• R CRAN SITE downloads, documentation, and more
• FAQs please read before posting a question! at CRAN or search using google
• Packages contributed packages for R
• R reference card (pdf) cheatsheet for useful functions
• Writing R extensions (pdf) Explanations of writing packages and calling C from R

---

JAGS

JAGS is Just Another Gibbs Sampler. It is a program for analysis of Bayesian hierarchical models using Markov Chain Monte Carlo (MCMC) simulation not wholly unlike BUGS. The name is a misnomer as JAGS implements more than just Gibbs Samplers. JAGS was written with three aims in mind:

• To have a cross-platform engine for the BUGS language
• To be extensible, allowing users to write their own functions, distributions and samplers.
• To be a plaftorm for experimentation with ideas in Bayesian modelling

Resources for JAGS:

• JAGS website on Sourceforge for Downloads, Manuals, etc
• JAGS News Martin Plummer’s
• site for all things JAGS
• rjags on CRAN

git

Git is a state-of-the-art version control system. It lets you track who made changes to what when and has options for easily updating a shared or public version of your code on github.

• OSX - install Git for Mac by downloading and running the installer or install homebrew and use it to install git via brew install git.

• Unix/Linux - you should be able to install git via your prefered package manager (if it is not already installed).

• Windows - install Git for Windows by download and running the git for windows installer. This will provide you with git, the bash shell, and ssh in windows.

• Screencasts from STA523 will be helpful in refreshing/learning how to use git and github