Introduction: How to study applied statistics with Python and R

This section of the documentation is inspired by the open textbook Applied Statistics with R by David Dalpiaz (University of Illinois at Urbana–Champaign), which is available under a Creative Commons Attribution–NonCommercial–ShareAlike 4.0 license.

PyStatsV1 adapts the spirit and core ideas of that text for a Python-first audience, while still keeping R in the conversation. The goal is not to replace the original book, but to give students and instructors a way to:

• see familiar R-based ideas expressed in plain Python code,

• run reproducible examples chapter by chapter,

• build intuition by comparing R and Python side by side.

Where the original text says “this book,” you can think of this documentation plus the PyStatsV1 codebase as our Python companion volume.

About this guide

This guide is meant to support:

• Students in an applied statistics course who are more comfortable in Python than in R.

• Instructors and TAs who are using an R-first textbook but want to demo the same ideas in Python.

• Practitioners who want a quick, “textbook style” reference for common applied methods implemented as transparent scripts.

The design philosophy is:

• Code first. Every idea should have runnable code in both languages.

• Reproducible by default. Synthetic data and fixed seeds make it easy to rerun examples and explore “what if” questions.

• Bridging R and Python, not replacing R. We treat R as a peer language, not a rival.

How this relates to the original R text

The original Applied Statistics with R book was designed for STAT 420 (Methods of Applied Statistics) at UIUC. It is still actively developed and is an excellent R resource.

PyStatsV1 builds on its structure and motivations, but:

• implements examples as plain Python scripts,

• adds reproducible CLI workflows (via make),

• and encourages reading the R and Python versions together.

A typical workflow might be:

1. Read a section from the original R text to understand the setup.

2. Run the corresponding PyStatsV1 chapter scripts to see the same ideas in Python.

3. Compare the R and Python output and code style.

4. Try small experiments: change the seed, sample size, or model and observe the effect.

Code conventions in this documentation

To keep things clear when switching between languages, we use consistent conventions.

Python code

Python code in the documentation appears in fenced blocks and matches the scripts in this repository. For example:

# Python example
import numpy as np

rng = np.random.default_rng(123)
x = rng.normal(loc=0, scale=1, size=100)
x.mean()

R code

Occasionally we will show R snippets for comparison. These will be clearly labeled and follow the usual R console style:

# R example
set.seed(123)
x <- rnorm(100, mean = 0, sd = 1)
mean(x)

When you see both versions together, the idea is:

• Same statistical idea, different syntax.

• Focus on the model and reasoning, not just the language.

Mathematical notation

As in the original text, we occasionally use symbols like \(p\) to denote the number of \(\beta\) parameters in a linear model. You do not need to memorize every symbol immediately; the important point is to connect:

• the model equation,

• the R implementation, and

• the Python implementation.

Where to report issues or suggest improvements

Just like the original book, this project is a work in progress. You may encounter:

• typos or unclear explanations,

• small discrepancies between R and Python output,

• places where the documentation could use another example.

If you do, we would love to hear from you.

• Use GitHub issues on the PyStatsV1 repository: https://github.com/pystatsv1/PyStatsV1/issues

• For general questions or teaching stories, use GitHub Discussions: https://github.com/pystatsv1/PyStatsV1/discussions

Helpful ways to contribute:

• Suggest rewording a confusing paragraph.

• Point out where the Python code could better match the R text.

• Propose a new small example or diagnostic plot.

• Submit a pull request if you are comfortable with Git and GitHub.

Acknowledgements and license

This guide owes a major intellectual debt to:

• David Dalpiaz, author of Applied Statistics with R.

• The STAT 420 teaching team and contributors acknowledged in the original text.

The original book is available at:

• Repository: https://github.com/daviddalpiaz/appliedstats

• PDF: http://daviddalpiaz.github.io/appliedstats/applied_statistics.pdf

Our adaptation for PyStatsV1 follows the Creative Commons Attribution–NonCommercial–ShareAlike 4.0 International License of the original work. In particular:

• You are free to share and adapt this material for non-commercial purposes.

• You must provide appropriate attribution to the original author.

• Derivative works must use the same license.

PyStatsV1 extends this by adding:

• Python implementations of textbook-style analyses,

• command-line workflows and CI tests,

• and documentation tailored to a Python + R learning environment.

Your contributions

If you contribute a substantial improvement to this documentation or the associated code and would like to be acknowledged, feel free to open an issue or pull request and indicate how you would like your name to appear. We are happy to recognize contributors and, if desired, link to a GitHub or personal website.

If you care about open, reproducible statistics education and want to learn why methods work by seeing them in both R and Python, you’re exactly the audience this guide was written for.