My Reproducible Data Analysis Project



This is a example of a data analysis project with focus on reproducibility using reprodown. This is a package that integrates:

  • Blogdown: An R package that integrates R markdown with hugo to create webpages.
  • Makefile: Automatise scripts execution with dependency. It only updates what is necessary.
  • scholar-docs: A custom hugo theme to show your executed scripts.

About this example

In this example, we provide scripts that create, clean and transform a dataset. Later, an exploratory analysis is provided. Classic and additive linear models are used to fit the data. A summary is provided in the last section.

Structure

Below you can see the structure of the project. reprodown reads the scripts folder to create the Makefile containing the recipes that define the dependencies between the files. Then the makefile creates the files required to create the website by Hugo.

.
├── data
│   ├── cleaned
│   ├── modelled
│   ├── processed
│   └── raw
├── docs
│   ├── config.toml
│   ├── content
│   ├── layouts
│   ├── resources
│   ├── static
│   └── themes
├── Makefile
├── scripts
│   ├── 10-extract
│   ├── 20-clean
│   ├── 30-process
│   ├── 40-explore
│   ├── 50-model
│   ├── 60-summarise
│   └── _index.Rmd
└── src
    └── 21-clean-group.R