Below is a small chart that outlines common options for statistics / econometrics software. gretl, EViews and MATLAB tend to have better-organised documentation. My personal choices:

  • if I had to choose just one: python;

  • if I had to choose other two: R (libraries) and gretl (documentation);

  • before open source era: Eviews;

  • if I had more time: Julia.

Open source


  • R (RStudio), derived from S

  • Python (Anaconda)

  • Julia

  • gretl

  • EViews (derived from TSP)

  • Octave


  • SAS

  • SPSS

  • Stata

RATS, Ox, PcGive

Stan, PyMC3, Turing.jl


SAS and terminology

SAS, it seems, has become the gold standard, the output of SAS programs the ultimate point of reference for correct and appropriate statistical calculations and the SAS terminology israpidly taking over as standard terminology. This is very Microsoft-like indeed and very worrying for anyone who cares about the profession.

WN Venables. Exegeses on Linear Models - on early adoption of S Plus

Example: software used in black hole discovery

Reading First M87 Event Horizon Telescope Results. III. Data Processing and Calibration: