GRETL is the GNU Regression, Econometrics, and Time Series Library. It has binaries for Debian, and also for Windows and OS X. The OS X versions comes nicely wrapped as a dmg, and installs as an double-clickable X11 program. GRETL interfaces through menu commands to R, to X12-ARIMA, and to TRAMO/SEATS.
The X-12-ARIMA seasonal adjustment program, with all of its documentation and utilities, is available from the Census Bureau. There are binaries for SUSE Linux and for Windows and there is FORTRAN source code, which compiles without problems using Intel’s ifort on OS X. I don’t know if it compiles using gfortran. There is also X12-GRAPH, with is a SAS program that provides a GUI for X-12-ARIMA.
X-12-ARIMA fits regARIMA models, i.e. regression models with ARIMA errors. There are build-in regressors for removing periodicities, but user defined regressiors are also possible.
TRAMO stands for Time Series Regression with ARIMA noise, Missing values and Outliers and SEATS for Signal Extraction in ARIMA Time Series. Programs are developed amd distributed by the Banco de Espana. There are DOS, Windows, and Linux binaries available, as well as a MATLAB interface. Also Excel utilities to create data files, and several papers with applications.
- SSA/MTM Toolkit
- Developed by the UCLA Department of Atmospheric and Oceanic Sciences. The program can do both Singular Spectrum Analysis and Multitaper Spectral Analysis. The GUI is X11/Motif based and runs on the various Unix and Linux versions, but there is a (commercial) native OS X version available from SpectraWorks. SSA is interesting, because it is exploratory and model free. The SSA/MTM website has links to many articles using the technique, mostly in geophysics and atmospheric sciences.
A competitor (commercial, for Windows, made in Russia) is the Caterpillar-SSA software, that implements techniques in the book
<a href=“http://www.gistatgroup.com/cat/book2/index.html”">Analysis of Time Series Structure. The Caterpillar website also has links to some interesting papers.
- A First Course on Time Series Analysis
- This is an open source book, which can be downloaded as pdf, and to which anyone can make contributions. It comes with a large number of SAS programs.
- Visual exploration of time series data. Java based interactive querying and exploring.