Tidy Finance Blog

Experimental and external contributions based on Tidy Finance with R. Contribute your ideas!

The legacy pandas edition of Tidy Finance with Python (unmaintained archive)

Python

Data

We migrated Tidy Finance with Python from pandas to polars and merged the R and Python editions into unified chapters. For reference, this post archives the complete rendered output of the previous pandas edition. It is no longer maintained.

Christoph Scheuch, Stefan Voigt, Patrick Weiss

High-Frequency Market Data: lobsteR and 20 Years of SPY on Hugging Face

R

Data

Market microstructure

Announcing lobsteR, an R package for LOBSTER high-frequency data, and seamless access to 20 years of 5-second SPY trading data via tidyfinance.

Tidy Finance migrates from SQLite to Parquet

Data

R

Python

Parquet dramatically improves performance and cross-programming language consistency.

Christoph Scheuch, Stefan Voigt, Patrick Weiss, Christoph Frey

ISS Shareholder Proposals

Data

R

Code for preparing ISS Voting Analytics data for further analysis on shareholder proposals

Alexander Pasler, Moritz Rodenkirchen

lobsteR package and high-frequency data

R

Data

Market microstructure

Introducing the lobsteR package and tidy cleaning procedures in R for LOBSTER high-frequency data

Replicating Gu, Kelly & Xiu (2020)

R

Replications

Machine learning

A partial replication of the paper Empirical Asset Pricing via Machine Learning using R.

CIR Model Calibration using Python

Interest rates

Python

Routine to calibrate the Cox-Ingersoll-Ross model

CRSP 2.0 Update

Data

R

Python

The highlights of the recent switch to CRSP 2.0 data

Patrick Weiss, Christoph Scheuch, Stefan Voigt, Christoph Frey

Tidy Market Microstructure

Market microstructure

R

data.table

A beginner’s guide to market quality measurement in high-frequency data using R.

Björn Hagströmer, Niklas Landsberg

Using DuckDB with WRDS Data

Data

R

Demonstrate the power of DuckDB and dbplyr with WRDS data.

Comparing Fama-French Three vs Five Factors

Data

Replications

R

An explanation for the difference in the size factors of Fama and French 3 and 5 factor data

Christoph Scheuch

Convert Raw TRACE Data to a Local SQLite Database

Data

R

An R code that converts TRACE files from FINRA into a SQLite for facilitated analysis and filtering

Kevin Riehl, Lukas Müller

Non-Standard Errors in Portfolio Sorts

Replications

R

An all-in-one implementation of non-standard errors in portfolio sorts

Construction of a Historical S&P 500 Total Return Index

Data

R

An approximation of total returns using Robert Shiller’s stock market data

Christoph Scheuch