sds (Statistics for Data Science)

sds collects the statistics that data science courses assume, as Quarto fragments:

• empirical distributions and descriptive statistics;
• estimation and statistical inference;
• maximum likelihood and Bayesian inference;
• basic statistical methods, such as t-tests, chi-square tests, and the bootstrap.

It renders on its own as a website, and course sites link to its pages by URL.

Using these notes in another site

Link to the pages by URL. A host site that keeps a copy of this repository at its root, named sds, can also include fragments with paths that start with sds/.

Quarto resolves @id cross-references only within one rendered page, so a host site that links to a result here uses an explicit link to the page and anchor.

Building the site

This repository uses the latex-macros submodule for its LaTeX macros, so clone it with submodules:

git clone --recurse-submodules https://github.com/Morrison-Lab/sds.git
quarto preview

The pages run R code. Install the R packages listed in DESCRIPTION (for example, pak::local_install_deps()), and the JAGS library, which the page of models fitted with JAGS uses.

Provenance

The notes began as the statistics appendices of Regression Models for Epidemiology, and keep the file names of those chapters, and most of their #id anchors; the home page lists the results that moved.

The site scaffolding comes from the qwt Quarto website template.