mattocci27 / stan-models

Stan examples for regression models

This repository contains examples demonstrating the use of Stan for regression modeling, with a focus on efficiency and reproducibility in data analysis workflows.

Requirements for running this demo

• R (>= 4.0.0)
• Quarto: An open-source scientific and technical publishing system built on Pandoc
• cmdstan: The shell interface to Stanj
• renv: A dependency management toolkit for R. This will install the key packages, including: targets, stantargets, tachetypes, cmdstanr, tinytex, and ensure the installation of cmdstan and TinyTex.

Usage

Forking the Repository and Creating a New Branch

To interact with this repository:

1. Fork the Repository:

   • Go to the original repository on GitHub.
   • Click the "Fork" button to create your copy.

2. Clone Your Fork:

   • Clone it to your machine: git clone [URL of your fork].

3. Create a New Branch:

   • Inside the cloned directory: git checkout -b [new_branch_name].

4. Stay Updated:

   • Set the original repository as "upstream" and regularly pull updates.

Benefits of Forking:

• Personal Exploration: Freely experiment with the code without affecting the main repository.

• Version Control: Practice using Git, a crucial skill in complex data analysis.

• Ongoing Updates: Easily merge updates from the main repository into your fork.

Restoring packages

Reinstall packages from renv.lock:

r$ |> renv::restore()

Running code on local

Execute using the provided script (run.sh):

> run.sh
1) tar_make() on local
2) tar_make_clustermq() on local
Enter number：

1. Run tar_make() locally for standard processing.
2. Use tar_make_clustermq() locally for parallel processing.

Exercises

Try exercises in docs/regression.qmd or docs/regression.html. These exercises focus on vectorization, reparameterization, and using targets and stantargets for efficient, reproducible workflows.