FSDA Toolbox extends MATLAB and statistics toolbox to support a robust and efficient analysis of complex data sets.
FSDA was first released on March 2012. The version released in February 2015 (v3.0) contains many news. We added new robust regression (tau) and multivariate estimators (Stahel-Donoho) together with new weight functions (hyperbolic, Hampel and optimal). We introduced routines for robust Bayesian regression and robust heteroskedastic regression. We added cluster analysis routines (tclust, tkmeans, MixSim) with related example datasets, and new exploratory tools for fininding the number of groups based on random start monitoring of minimum deletion residual. We also integrated routines for connecting MATLAB with R. There is a new function to obtain proper threshold for robust estimators in order to have an empirical size equal to the nominal (function RobRegrSize.m). We added univarite robust scale estimators Qn and Sn. There is a new function to compute the cdf of the linear combination of non central chi-squared random variables (ncx2mixtcdf). Among the new utilities, an efficient function to extract elements above diagonal in a vector (function triu2vec.m) and to position figures (function upperfracpos.m).
In the Installation Note Section or pdf file we tried to document all that you have to expect when FSDA is installed manually by unpacking the compressed tar file FSDA.tar.gz, or automatically with our setup program for Windows platforms.
DATE OF LAST UPDATE: February 10, 2015
Download FSDA A setup executable for MS Windows platforms will install the toolbox and update the search path of your local MATLAB installation.
Download a compressed tar file of the toolbox, suitable for Unix platforms installation. In this case, you have to add manually FSDA folder and sub-folders to the MATLAB path or use our routine addFSDA2path.
Download a working paper describing the main characteristics of the FSDA toolbox or see RIANI, M., PERROTTA D., TORTI F. (2012).
You can have a look at the toolbox features through some quick screenshots or through some didactic movies
Movie 1: Hawkins data
Movie 2: Fishery data
Movie 3: AR data
Movie 4: Loyalty cards data
Movie 5: Hospital data
Please, do not hesitate to contact us for any bug you might find and for any suggestion you might have!!! Thanks to all those who have already contacted us and have helped us to correct several bugs and improve the performance of the code.
Marco Riani, Andrea Cerioli and Aldo Corbellini (University of Parma).
Domenico Perrotta and Francesca Torti (EC, Joint Research Centre).
Patrizia Calcaterra, Emmanuele Sordini, Daniele Palermo (EC, Joint Research Centre).