Gnu Regression, Econometrics and Time-series Library
Is a cross-platform software package for econometric analysis, written in the C programming language. It is free, open-source software. You may redistribute it and/or modify it under the terms of the GNU General Public License (GPL) as published by the Free Software Foundation.
gretl conference 2013
The third biennial gretl conference took place in Oklahoma City on 20-21 June, 2013; details here. Many thanks to Lee Adkins for organizing the event! At this point it looks as if our next conference (in 2015) will take place in Berlin.
Supported formats include: own XML data files; Comma Separated Values; Excel, Gnumeric and Open Document worksheets; Stata .dta files; SPSS .sav files; Eviews workfiles; JMulTi data files; own format binary databases (allowing mixed data frequencies and series lengths), RATS 4 databases and PC-Give databases. Includes a sample US macro database. See also the gretl data page.
Whether you're compiling from source or installing a pre-compiled binary, you may want/need to check out gretl's dependencies.
The time-series analysis programs X-12-ARIMA and TRAMO/SEATS are available in a form suitable for use with gretl on Linux, as follows.
Gretl calls gnuplot to generate graphs. We require gnuplot version 4.4.0 or higher, but version 4.6.5 (the current release as of May 2014) or higher is recommended. See gnuplot.sourceforge.net. The gretl packages for MS Windows and OS X include a gnuplot version that works well with gretl. For guidance on installing current gnuplot on Linux, please look here.
The gretl manual comes in two PDF files, a User's Guide and a Command Reference. English versions of these, formatted for U.S. letter-size paper, are included in the gretl source package and binary distributions. Gretl will access the other available versions -- namely English (A4 paper), Italian and Spanish -- on demand, via the Internet. You can also find the manual files here.
If so you may wish to join the gretl-users mailing list. This is a moderate-volume list where people request and offer help, discuss new features and so on. The list archive is searchable. There's also a gretl-announce list; this is a read-only list which sends out announcements on the occasion of new releases of gretl.
You can submit bug reports and feature requests using the SourceForge Tracker system. But please note: we request that you use the bug tracker only if you're fairly sure you've really found a bug; if you're uncertain, please send a message to one of the gretl mailing lists first. (Needless to say, if gretl crashes you've really found a bug alright!)
Do you have ideas or code that you'd like to contribute to gretl? If so, we'd be very pleased to hear from you. You should probably join the gretl-devel mailing list (this is fairly low volume).
The only requirement on code that can be contributed to gretl is that you must be willing to have it appear under the GNU General Public License (GPL), that is, you must be comfortable with having the source released.
As mentioned above, gretl is (mostly) written in the C programming language, the most common language in the realm of free, open-source software. However, not everyone is comfortable writing in C. So you should be aware that this need not be a barrier to contributing code or algorithms. If you can supply a well specified algorithm to accomplish some interesting econometric task, we can probably either interface with it or rewrite it in C without too much difficulty. Possibilities include code in Fortran or C++; code written in a high-level matrix manipulation language such as Ox, Gauss, Octave, Matlab or R; and even pseudo-code, that is an algorithm spelled out explicitly, step by step, but not in any particular programming language.
First of all, thanks to Ramu Ramanathan, Professor Emeritus of the University of California, San Diego, for open-sourcing his "ESL" econometric code, which was the starting point for the development of gretl. Professor Ramanathan is author of Introductory Econometrics (Dryden, currently in its 5th edition). Ramu has also been a very helpful critic over the course of gretl's development.
Many people have sent in useful bug reports and suggestions for gretl's development. We are particularly indebted to Ignacio Díaz-Emparanza, Tadeusz Kufel, Pawel Kufel, Dirk Eddelbuettel, Sven Schreiber and Andreas Rosenblad. A. Talha Yalta plays a helpful role in scrutinizing and reporting on gretl's numerical accuracy.
Many thanks to Ignacio Díaz-Emparanza, Susan Orbe, Michel Robitaille, Florent Bresson, Cristian Rigamonti, Tadeusz and Pawel Kufel, Markus Hahn, Sven Schreiber, Hélio Guilherme, Henrique Andrade, Alexander Gedranovich, Talha Yalta, Y. N. Yang, Pavla Nikolovova, Jan Hanousek, Artur Bala, Manolis Tzagarakis and Ioannis Venetis for their work in translating gretl.
Thanks to William Greene, author of Econometric Analysis, for his permission to include in the gretl package some datasets relating to interesting examples in his text.
Thanks to the good people on comp.lang.c and firstname.lastname@example.org for expert advice on many issues. Thanks to Richard Stallman of the Free Software Foundation for all his work in developing and promoting free software, and more specifically for agreeing to "adopt" gretl as a GNU program.
Allin Cottrell, Wake Forest University
Riccardo "Jack" Lucchetti, Università Politecnica delle Marche
Last modified: 2014-05-05