📊 Learning Stan

Welcome! This repo documents Alon Kagan's journey learning Stan and Bayesian statistics in general.

📘 Learning Resources

I'm following this guide:

• Stan Getting Started Guide by Bob Carpenter

Additionally, I will try to recreate results from an unpublished paper (kept private for now).

Lastly, there is probably a youtube course worth watching, here (if I have spare time)

---

⚙️ Installation Instructions

1. Install WSL (Windows Subsystem for Linux)

If you're using WSL for the first time, update and install required packages:

sudo apt update && sudo apt upgrade -y
sudo apt install -y python3 python3-pip build-essential git wget unzip
sudo apt install -y python3-venv

2. Set Up Python Virtual Environment for CmdStanPy

python3 -m venv ~/cmdstanpy-env
source ~/cmdstanpy-env/bin/activate

3. Upgrade Pip

pip install --upgrade pip

4. Install Required Python Packages

Make sure you have a requirements.txt file in your directory, then run:

pip install -r requirements.txt

5. Install CmdStan via CmdStanPy

python -c "import cmdstanpy; cmdstanpy.install_cmdstan()"

---

✅ You're All Set! From now on, whenever you want to use CmdStanPy, just:

source ~/cmdstanpy-env/bin/activate

And you're good to go!

Usage

open a new session:

wsl
source ~/cmdstanpy-env/bin/activate

🧠 Author

Alon Kagan — Exploring Bayesian inference and Stan one model at a time. (This joke was made by ChatGPT when I told it to make this page look nicer)