★ Open-source · R & Python · Reproducible
Empirical finance, reproducible in R and Python
Tidy Finance takes an opinionated, fully transparent approach to empirical research in financial economics. One open-source code base now covers R and Python, side by side in every chapter.
library(tidyfinance)
library(ggplot2)
prices <- download_data(
domain = "Stock Prices",
symbols = "AAPL",
start_date = "2000-01-01",
end_date = "2024-12-31"
)
prices |>
ggplot(aes(date, adjusted_close)) +
geom_line()import tidyfinance as tf
from plotnine import ggplot, aes, geom_line
prices = tf.download_data(
domain="Stock Prices",
symbols="AAPL",
start_date="2000-01-01",
end_date="2024-12-31"
)
(ggplot(prices, aes("date", "adjusted_close"))
+ geom_line())20+ Hands-on chapters
2 Languages, one code base
100% Open source & reproducible
Everything you need for reproducible research in finance
A single, transparent code base takes you from raw data to publication-quality results, in R and Python.
One code base, two languages
Every chapter ships R and Python side by side from a single source. Switch with a tab. Your choice follows you through the book.
Fully transparent & reproducible
Open code you can run end to end, from raw data to every figure and table, with no black boxes.
Powered by tidyfinance
A companion package for R and Python that wraps data downloads and the routines we use throughout the book.
Real financial data
Work with the data professionals use, like CRSP, Compustat, and TRACE, through clean, documented access patterns.
From sorts to machine learning
Portfolio sorts, factor models, fixed effects, causal inference, and modern ML give you the full empirical toolkit.
A free, open book
Read the whole book online for free or dive into the blog all in the open.
Browse the book by topic
Five parts take you from your first stock return to constrained portfolio backtests.
Getting Started
Stock returns, modern portfolio theory, the CAPM, and financial statement analysis.
Financial Data
Access and manage WRDS, CRSP, Compustat, TRACE, and FISD with documented workflows.
Asset Pricing
Beta estimation, univariate and bivariate sorts, Fama-French factors, and Fama-MacBeth.
Modeling & Machine Learning
Fixed effects, difference-in-differences, factor selection, and option pricing via ML.
Portfolio Optimization
Parametric portfolio policies and constrained optimization with realistic backtesting.
Praise
What experts say about Tidy Finance
A clean coding environment is a prerequisite for building a relevant investment platform and conducting meaningful factor research. Tidy Finance is the name of the game, giving aspiring academics and finance practitioners just what they need to perform clean and reproducible research. Highly recommended.
From our crowd-sourced paper on non-standard errors, I learned how important clean coding is. Tidy Finance is a rich resource for empirical finance researchers, offering clean coding techniques that benefit both beginners and experts.
A fantastic book bringing together financial theory, sound econometrics, thorough data processing and powerful programming techniques. An absolute must for every student and scholar in empirical finance.
Tidy Finance is a fantastic resource that lowers the threshold for entry into empirical finance, all in the spirit of open and reproducible science.
To have a deep understanding of empirical asset pricing, one needs to write code using actual data. To learn how to do this, there is no better starting point than Tidy Finance. [...] I strongly recommend Tidy Finance to both beginners and experts.
Students and professionals alike are led step by step until they suddenly find themselves coding on their own. A brilliant and required resource!
The team
Who maintains Tidy Finance
Christoph Scheuch
Associate Researcher at HU Berlin
LinkedIn →
Stefan Voigt
Assistant Professor of Finance at University of Copenhagen
Website →
Patrick Weiss
Assistant Professor of Finance at Reykjavik University
Website →
Christoph Frey
Quantitative Researcher
Website →