Software for the Mac Banner



by A. Antoniadis , J. Berruyer and A. Filhol

Current Version: 4.1
68K and PowerPC
Updated: 1/7/97

ABCstat is a program implementing various statistical tools on the Macintosh. It will run on any Macintosh with at least 8 MB of RAM and System 7. Here are some files that ABCstats potential users may find useful:




by A. Antoniadis , J. Berruyer and A. Filhol

Commercial - Academic discount available.
Current Version: 2.0
68K and PowerPC
Updated: 12/18/96

MacSurvival is a program implementing Survival Analysis on the Macintosh. It will run on any Macintosh with at least 8 MB of RAM and System 7. It is, we believe, the only Survival Analysis implementation for the Macintosh to include a nonparametric hazard smoothing module.




 by F. R. Degott and O. Gaudoin

Version: 2.0
68K and PowerPC
Last update : 22/3/94

LogiFiab incorporates the main probabilistic models and tests statistics for evaluating software reliability. It can be used in the following domains:

  • industry : for the reliability evaluation of software,
  • research : for experimenting new techniques,
  • teaching .

LogiFiab can be used on any type of Macintosh. There are two versions : with ou without arithmetic processor.




Fractional and Multifractional

Brownian Motion


by Jean-Francois Coeurjolly

Version: 1.0
Last update : 22/06/01

In this .tar compressed archive various methods for generating a discretized path of a fractional and mutltifractional Brownian motion and for estimating its parameters are implemented and/or discussed in detail in S-plus and/or Matlab. To download the S-plus code and the accompanying paper, related to the problem of simulation and identification of the fractional Brownian motion, click on the following link:

To download the Matlab code about the simulation and identification of both processes (procedures supporting the PhD Thesis of J.-F. Coeurjolly (2000)), click on the following link:




Gaussian Wavelet Denoising Matlab Toolbox


Various wavelet shrinkage and wavelet thresholding estimators in nonparametric regression appearing in the literature are discussed in detail and implemented in MATLAB. These estimators arise from a wide range of classical and empirical Bayes methods treating either individual or blocks of wavelet coefficients.To download the code or the accompanying paper, click on the following link:

  Smoothing Toolbox

The Smoothing Toolbox is a flexible and extensible collection of Matlab functions, scripts, investigating smoothers and their applications. The main aim of these routines is research -- to develop specific tools for specific goals in Nonparametric Data Analysis. The software is free to anyone interested enough to download it, however to run the software you will of course need MATLAB. All I ask in return is that you

  • acknowledge use of this material (if you find a use for it);
  • let me know via email any bugs or suggestions for improvement.

I would like to thank B. Seifert, I. Gijbels, J. Fan and D. T. Pham for providing me the necessary sources for implementing the main functions in this toolbox. Special thanks to B. Seifert for his careful reading and testing of several of those scripts and for making them also available to the community through the web server of his University.

Anestis Antoniadis.

(Click here for a French version)


Empirical Bayes Thresholding

This software complements the paper "Needles and straw in haystacks: Empirical Bayes approaches to thresholding a possibly sparse sequence" and "Empirical Bayes selection of wavelet thresholds" by Iain M. Johnstone and Bernard W. Silverman, submitted for publication 2002. A paper giving a general description of the software and some details both of the general methodology and of some specific technical matters is available here.

(Click here for a French version)


[ Main Page | Members | Preprints | Projects | Publications | Research | Seminars | Software | LJK | IMAG ]