install-R-packages.Rmd

---
title: "Script for installing R packages for Stat4Astro School, Autrans"
author: "Julyan Arbel"
output: 
  rmarkdown::html_document:
    toc: true
    toc_float: true
    toc_depth: 3
    number_sections: true
---


# Purpose
This script is to provide the participants of the [School of Statistics for Astrophysics 2017: Bayesian Methodology](https://stat4astro2017.sciencesconf.org/) with the necessary R packages.


# Prerequisistes

Have [R](http://cran.r-project.org/) installed, as well as [RStudio](https://www.rstudio.com/), Desktop version.

# Access to Slides

The folder contains the slides for the following presentations:

  * David van Dyk
  * Roberto Trotta
  * Christian Robert (TBA)
  * Julyan Arbel (TBA)

# Installing R packages

In order to install the required R packages, you can either:

- run the following lines in RStudio: copy-paste them to RStudio console (bottom-left panel)

- or open this entire document in RStudio and run it (alt-cmd-R on a Mac).

```{r}
# General R packages (visualization)
install.packages("ggplot2", "rgl","rglwidget", "gridExtra", "hexbin", "viridis")

# MCMC & Model Building
install.packages(c("mvtnorm","coda"))

# Approximate Bayesian Computation (Christian Robert)
install.packages(c("abcrf","abc"))

# Bayesian Nonparametrics (Julyan Arbel)
install.packages(c("DPpackage","BNPdensity","mcclust"))
# the package mcclust.ext is not on CRAN and requires the specific install from tar.gz file as described below
download.file(url = "http://wrap.warwick.ac.uk/71934/1/mcclust.ext_1.0.tar.gz", 
              destfile = "mcclust.ext_1.0.tar.gz")
install.packages("mcclust.ext_1.0.tar.gz", repos = NULL, type = "source")
file.remove("mcclust.ext_1.0.tar.gz")
```