importr-stata (v2.0.3)

A dual-bridge (R/Python) Stata command for importing R data files (.Rdata, .Rda, .Rds).

---

🚀 Overview

importr v2.0 is a major upgrade that provides a robust way to bring R datasets into Stata. It is designed to work even if R is not installed on the system, by falling back to Stata's native Python integration. It is not practically possible to import R data using "pure" Stata commands (like file read or Mata) because .Rdata and .RDS files are complex, compressed binary objects that use a custom serialization format. Parsing them from scratch in Stata would require building a full binary decoder in Mata, which is a massive engineering undertaking. To avoid this, the "Hybrid" approach in importr first looks for R installed, if it doesn't find R, it will attempt to use Stata's built-in Python integration (available in Stata 16 and newer). This provides two paths to success:

1. Path A (R): Uses Rscript and the haven package (best for R users).
2. Path B (Python): Uses Stata's internal Python with the pyreadr library (to read the R file) and pyreadstat (to write the .dta) (best for modern Stata environments).

✨ Features

• Dual-Bridge Architecture:
  • Primary (R): Uses Rscript and the haven package.
  • Secondary (Python): Uses Stata's python integration with the pyreadr and pyreadstat libraries.
• Format Support: Works with .Rdata, .Rda, and .Rds.
• Intelligent Fallback: Automatically detects the best available tool on your machine.

---

🛠️ Installation & Requirements

Install from GitHub in Stata:

net install importr, from("https://raw.githubusercontent.com/ericabooth/importR-stata/main/") replace force
discard
which importr
help importr

To pull the example R datasets alongside the command, net get the ancillary files:

net get importr, from("https://raw.githubusercontent.com/ericabooth/importR-stata/main/")
importr using importr_example.rdata, clear
importr using importr_example.rds, clear

net get drops the two example files, importr_example.rdata and importr_example.rds, into the current directory.

importr needs one of the two bridges below on the machine.

For the R Bridge (Default)

• R and Rscript must be in your system's PATH.
• The R package haven. If it is missing, the bridge installs it from CRAN (install.packages("haven")) before converting, which needs an internet connection and writes to your R library.

For the Python Bridge (Fallback)

• Stata 16 or newer.
• The Python libraries pyreadr, pyreadstat and pandas (pyreadr will not import without pandas), installed into the Python that python query reports. python pip is not a Stata command, so run pip from a terminal:

  /path/to/that/python3 -m pip install pyreadr pyreadstat pandas

  When the Python bridge cannot run, importr prints this command with the right path.

---

📖 Quick Start

* The command will automatically choose the best method
importr using "data/my_results.Rdata", clear

* Works with RDS files too (detected from the extension; the rds option forces it)
importr using "data/model_fit.rds", clear
importr using "data/model_fit.dat", clear rds

* What was read
display r(nobs) " obs, " r(nvars) " vars from " r(filename) " (object " r(object) ")"

Options: clear replaces the data in memory (without it, changed data in memory stop importr with r(4)); rds treats the file as an RDS file.

Stored results: r(nobs), r(nvars), r(filename), and, for .Rdata/.Rda files, r(object), the name of the object read. A file with several objects gives the first data frame by name (C-locale order), the same with either bridge.

📤 Bonus Utilities: Python Standalone Tools

While importr is designed for Stata users, this repository also includes standalone Python scripts for converting between Stata and R formats outside of Stata.

Requirements

Ensure you have the required Python libraries:

pip install pyreadstat pyreadr pandas

1. RtoStata.py (R → Stata)

Convert R data files (.rds, .Rdata) to Stata .dta format:

# Convert to DTA (default)
python RtoStata.py mydata.rds

# Specify Stata version (e.g., version 13)
python RtoStata.py mydata.Rdata -v 13

2. pythontoR.py (Stata → R)

Convert Stata .dta files into R formats (.rds or .rdata):

# Convert to RDS (default)
python pythontoR.py mydata.dta

# Convert to RData
python pythontoR.py mydata.dta --format rdata

---

Changelog

• 2.0.3 (9 Nov 2026): changed data in memory without clear now give r(4) instead of a misleading "needs R or pyreadr" error; the install advice prints a pip command to run in a terminal (python pip is not a Stata command); both bridges take the same object from a multi-object .Rdata file and report it in r(object). RtoStata.py now uses pyreadr.read_r() (it called read_rds()/read_rdata(), which pyreadr does not provide).
• 2.0.2 (Aug 2026): bridge subroutines are rclass; each bridge is attempted in turn; Windows paths.

👤 Author

Eric A. Booth

• 🏛️ Sr Researcher, Texas 2036
• 📧 eric.a.booth@gmail.com
• 🌐 www.eric-booth.com
• 💼 GitHub Profile

📄 License

This project is licensed under the MIT License.