SSJ: Stochastic Simulation in Java
What is SSJ?
SSJ is a Java library for stochastic simulation, developed
under the direction of Pierre L'Ecuyer, in the
Département d'Informatique et
de Recherche
Opérationnelle (DIRO), at the Université de Montréal.
It provides facilities for generating uniform and nonuniform random
variates, computing different measures related to probability
distributions, performing goodness-of-fit tests, applying
quasi-Monte Carlo
methods, collecting (elementary) statistics,
and programming discrete-event simulations
with both events and processes.
Documentation
- Official SSJ-1.2.5 documentation in
PDF format; it is more
detailed and more complete than the HTML documentation.
- SSJ-1.2.5 documentation in HTML
format.
It was translated from Latex to HTML by an automatic translator and,
as a consequence,
mathematical formulas are not always translated
correctly. However, the PDF documentation is translated correctly from
Latex.
- Examples programs with comments using SSJ:
- Parameter files for digital nets in package
hups, for class
DigitalNetFromFile.
- Parameter files for point sets in package
hups, for class
F2wStructure.
Downloading and Installing SSJ
WARNING: THIS IS AN OLD VERSION OF SSJ COMPATIBLE
WITH JDK-1.2. There will be no new developments on this version,
only bugs correction.
For a more modern version with added classes and packages,
and compatible with JDK-1.5, see
here.
Version: SSJ-1.2.5.
Compatible with JDK-1.2.
Last update: 17 may 2008.
SSJ is packaged in a ZIP file:
download SSJ.
``UnsatisfiedLinkError'' with the Chrono?
Read this note
Licence
SSJ was started several years ago by
Pierre L'Ecuyer and is copyrighted by him. A few coworkers and
students have participated in the development of SSJ:
their names may
be found in the
overview of the documentation.
SSJ may be used freely for personnal use, for research purposes or
in a non-commercial environment.
For use in a commercial environment,
please communicate with
Pierre L'Ecuyer.
Contact us
If you find a bug or if you want to suggest improvements, send mail to
Richard
Simard.
e-mail: simardr@iro.umontreal.ca