Release notes

This table provides quick access to what's new in each version.

FSDA follows typical MATLAB timetable in the sense that there are two releases per year. The first typically in May/June and the second around October/November.

Version (Release) New Features and Changes Version Compatibility Considerations Fixed Bugs and Known Problems Release dates
Function tclustreg now includes trimmed Cluster Weighted Restricted Models.

New function tclustIC for the automatic selection of the best number of groups.

New function tlclustICsol to extract a set of relevant solutions (and associated tclustICplot).

New function UnitsSameCluster to to control the labels of the clusters which contain predefined units.

New function to compare two partitions (Fowlkes and Mallows index)

Updated routine simdatasetreg to generate new outlier patterns


New routines for density estimation and thinning, for univariate and bivariate data (used in tclustreg).

bve, rthin, wthin, WNChygepdf

New functions wraptextFS, removeextraspacesLF

New functions mveeda, MMmulteda
Smulteda, Sregeda

 New function ScoreYJ which implements the score test for the Yeo and Johnson transformation. This new transformation has also been embedded inside function FSRfan.

SAMPLING AND COMBINATORIAL updated function randsampleFS and subsets.

NEW mlx files
.mlx files introduced for examples_MixSim

New function to create the car-bike plot to find the most relevant solutions.

Functions resfwdplot, malfwdplot generalized in order to take as input the output of procedures which monitor robust estimators.

  Control version added in many graphical routines to take account the modifications in MATLAB 2017a

From MATLAB R2012a callback functions started to use new internal data structures. Interactive plots that use such callbacks (e.g. those in GUI brushRES) may produce errors in releases older than R2012a. The issue should be now fixed. Please report to us any problem you may experience.

May 2017
V3.3 (R2016b) New functions for  bivariate density estimation and random thinning (kdebiv.m, rthin.m) used to extend tclustreg.m features.

The FSDA help folder now contains XML files associated to the functions documentation. This is in view of generating/updating automatically or using a GUI the functions documentation, in html as well as in the function head.

New html documentation generated with publishFS.
We are monitoring  possible compatibility issues that may emerge from  changes in the tagging policy of the graphical objects in MATLAB.
add2spm modified to take into account a change in the property name of the legend object introduced in MATLAB R2016b
(LegendPeerHandle is now called LayoutPeers).

A change in MATLAB R2016b  function legend.m was affecting FSDA function add2yX (the legend was plotted twice). Bug fixed.

Bug affecting FSMeda only in the univariate case fixed.
October 2016
V3.2 (R2016a) Major release.

New function, publishFS, introduced to generate documentation pages directly from the .m files.

New function, makecontentsfileFS, introduced to create a the list of files present in a FSDA folder and/or subfolders. It extends MATLAB function makecontentfilesFS.

.mlx files introduced for examples_multivariate and examples_regression.

New features added to the tclust function, including determinant restriction and new adjusted BIC criterion for the estimation of the number of groups.

Added functions for reweighting FSR and FSRB (FSRr and FSRBr).

Functions FSR, FSRB and FSRH redesigned; a routing implementing the core of the Forward Search algorithm (FSRcore) introduced to avoid code redundancies.

New function, winsor, to winsor data.

New function FSMbsb, which will replace FSMbbm.

New function randindexFS, to evaluate the quality of different clusterings.

New routines poolClose and poolPrepare introduced to conveniently open and close a pool of parallel workers.

Several new robust functions to generate, for example, the Tukey Biweigh rho function (HUrho),  the tuning constant associated to a certain efficiency (HUeff), the psi functions (HUpsi), its derivative (HUpsider), etc. For a full list, see functions under utilities_stats folder.

Documentation made compatible with the new MATLAB help navigation and browser-style, which was redesigned in release R2015b.

Many small bugs fixed. Functions affected include tclust, tclustReg, position, MixSimReg.

ClickableMultiLegend made compatible with R2016a.

drawnow command introduced in several graphical functions, to solve episodic bugs in the generation of the plots.

May 2016

V3.1 (R2015b)

Major release.

Function simdataset.m modified to allow the user to simulate outliers from different distributions and contamination schemes and/or contaminate existing datasets.

New Bayesian regression analysis routines: FSRB.m, FSRBeda.m, FSRBmdr.m, regressB.m.

In FSReda.m: monitoring of confidence intervals of beta and sigma2.

In FSRBeda.m: monitoring of HPD (highest posterior density regions) of beta and sigma2.

New functions for inverse gamma computation: inversegampdf.m, inversegamcdf.m, inversegaminv.m.

Added functions to monitor units forming subset in heterosckedastic and Bayesian regression: FSRHbsb.m, FSRBbsb.m.

Added new datasets for Bayesian examples.

FSMtra.m: added option for the robust transformation in the Yeo-Johnson family.

addFSDA2path.m: modified for compatibility with unix platforms and to address changes in the folder organization of FSDA functions.

Added routines to compute and visualize robust bivariate boxplot (function boxplotb.m)

New routine for automatic outlier detection in heteroskdastic regression (FSRH.m)

In function spmplot, the multiple histograms (boxplots) on the main diagonal of the scatter plot matrix is now working also with R2015a and R2015b.

The html documentation pages of some new functions are missing. As a consequence, some links may be broken. New complete documentation, produced automatically from the documentation in the head of our m-files, will be released soon.

September 2015

V3.0 (R2015a)

Major release.

Added robust regression (tau) and multivariate estimators (Stahel-Donoho). Functions taureg.m and SDest.m. New weight functions (hyperbolic, Hampel and optimal).
New routine for robust Bayesian regression (i.e. FSRBmdr).
New routine for robust heteroskedastic regression (i.e. FSRHmdr).
Added cluster analysis routines (tclust, tkmeans, MixSim), with related example datasets. New exploratory tools for fininding the number of groups based on random start monitoring of minimum deletion residual (fsrmdrRS.m, first FSDA function which exploits the parallel processing toolbox, if present).
New routines for connecting MATLAB with R.
New function to obtain proper threshold for robust estimators in order to have an empirical size equal to the nominal (function RobRegrSize.m).
Added univariate robust scale estimators Qn and Sn (functions Qn.m and Sn.m).
New utilities to extract elements above diagonal in a vector (function triu2vec.m) and to position figures (function upperfracpos.m).
New function to compute the cdf of the linear combination of non central chi-squared random variables (ncx2mixtcdf).

Documentation made compatible with the new help system introduced with MATLAB R2015a

In function spmplot, the multiple histograms (boxplots) on the main diagonal of the scatter plot matrix do not work with R2015a because of recent changes in gplotmatrix.

February 2015


V2.1 (R2013a)

New very fast implementation of the Forward Search for multivariate analysis (FSMmmd). Forward Search in regression modified to deal with the cases in which in a particular step of the search subset is not full rank (FSR and FSRmdr). New scatter plot matrix with multiple groups and multiple boxplots on the main diagonal (spmplot).

Structure modified for compatibility with MATLAB R2012b to R2013a releases. Three APPs introduced in release R2013a.


May 2013


V2.0 (R2011b)

Major release. Traditional robust estimators added including S, MM, MCD, MVE, (Sreg, Smult, MMreg, MMmult, mcd, mve) and univariate and bivariate analysis (unibiv). New interactive plots and interactive graphics features.

Redesigned the definition of many optional input parameters in terms of structures


September 2011


V1.1 (R2010b)

Multivariate data analysis routines have been added. Function resfwdplot considerably improved.



September 2010


V1.0 (R2010a)


A few functions use ~ to denote unused output parameters as from release 2009b


February 2010