gretl
gretl.sourceforge.net
- Free plan
- Windows
- macOS
- Linux
What gretl is
gretl is a free, open-source desktop package for econometric analysis on Linux, macOS, and Windows. Its methods include least squares, maximum likelihood, GMM, and regularized regression such as LASSO, Ridge, and elastic net. For time series, it offers ARIMA and GARCH-type models, VARs and VECMs, unit-root and cointegration tests, and the Kalman filter. It also includes panel-data estimators and limited dependent-variable models such as probit. Users can script in hansl and produce model output as LaTeX tables or equations. Data import covers CSV, Excel, OpenDocument, Stata, SPSS, and several other formats. gretl can exchange data and results with R, Octave, Python, Julia, Ox, and Stata; it also supports machine learning through LIBSVM and mixed time-series frequencies via MIDAS. The GNU GPL permits redistribution and modification. The maker describes gretl as user-friendly and sophisticated, while noting that its statistical-method coverage is less comprehensive than GNU R’s.
Who it suits
gretl suits students, researchers, and analysts who need econometric models, statistical tests, and scripting across desktop operating systems. It may be a fit for users working with supported data formats or exchanging results with other listed tools.
Good to know, and what to watch
- Available for Linux, macOS, and Windows.
- Includes time-series, panel-data, and regression methods.
- Hansl scripting and LaTeX model output.
- Imports data from CSV, Excel, Stata, and SPSS.
- GNU GPL allows redistribution and modification.
- Statistical-method coverage is less comprehensive than GNU R’s.
- macOS packages require macOS 10.9 or higher; Apple Silicon needs macOS 11 or higher.
Plans
Every plan and pricegretl
Free plan
The verdict
gretl brings a broad set of econometric methods, scripting, and data exchange to a free desktop package. Its stated method coverage is less extensive than GNU R’s, which is worth weighing when choosing a tool.
What the maker publishes
- Bayesian analysis
- Yes source
- Data formats
- Supported formats include CSV, Excel, Gnumeric, Open Document, Stata, SPSS, EViews, JMulTi, RATS 4 and PC-Give files, as well as gretl's own formats. source
- Data import formats
- XML, CSV, Excel, Gnumeric, OpenDocument, Stata, SPSS, EViews, JMulTi, RATS 4 and PC-Give source
- Deployment
- desktop source
- Estimators
- It includes least squares, maximum likelihood, GMM, single-equation and system methods, and regularized least squares methods including LASSO, Ridge and elastic net. source
- Free plan
- Yes source
- Integrations
- The site lists data and results exchange with GNU R, GNU Octave, Python, Julia, Ox and Stata, plus parallelization via MPI. source
- Intended users
- The maker describes gretl as a free, user-friendly and sophisticated econometrics package, and notes that its statistical-method coverage is less comprehensive than GNU R's. source
- Machine learning
- The site states that gretl supports machine learning via LIBSVM and mixed time-series frequencies via MIDAS. source
- macOS requirements
- The macOS download page offers gretl for macOS 10.9 or higher, with separate packages listed for Apple Silicon on macOS 11 or higher and for Intel Macs. source
- Other models
- It supports limited dependent-variable models and panel-data estimators, including instrumental variables, probit and dynamic panel models based on GMM. source
- Platforms
- The download page links to packages for Windows and macOS, and Debian and Fedora packages for Linux; it also provides source packages for compiling. source
- Purpose
- gretl is a cross-platform software package for econometric analysis, written in C. source
- Regression models
- Yes source
- Scripting and output
- gretl has an integrated scripting language called hansl and can output models as LaTeX files in tabular or equation format. source
- Scripting support
- Yes source
- Security and license
- The maker describes gretl as free, open-source software that may be redistributed and modified under the GNU General Public License. source
- Statistical tests
- Yes source
- Support
- Users can seek help through the gretl-users mailing list and submit bug reports and feature requests through the SourceForge Tracker system. source
- Time series
- Time-series methods include ARIMA, GARCH-type models, VARs and VECMs, unit-root and cointegration tests, and the Kalman filter. source
Checked against the maker's pages on 3 Oct 2026.