Scilab Home Page | Wiki | Bug Tracker | Forge | Mailing List Archives | Scilab Online Help | File Exchange
ATOMS : Stixbox details
Login with GitLab


Statistics toolbox
(6579 downloads for this version - 65293 downloads for all versions)
A more recent valid version with binaries for Scilab 5.4 exists: 2.4
Anders Holtsberg
Paris Sud University
Maurice Goursat
Jean-Philippe Chancelier
Serge Steer
Allan Cornet
Michael Baudin
Owner Organization
INRIA,DIGITEO, Universite Paris Sud, and others
Michael BAUDIN
Creation Date
March 14, 2013
Source created on
Scilab 5.4.x
Binaries available on
Scilab 5.4.x:
Windows 64-bit Windows 32-bit Linux 64-bit Linux 32-bit macOS
Install command
--> atomsInstall("stixbox")

Stixbox is a statistics toolbox which provides distribution functions, datasets,
statistical tests and plotting facilities. 


 * Cumulated Distribution Functions
   * pks : Kolmogorov Smirnov distribution function
 * Datasets
   * getdata : Returns a dataset.
 * Graphics
   * stixbox_graphics : Demos of the graphics.
   * bubblechart : Plot a bubble chart
   * bubblematrix : Plot a bubble chart matrix
   * histo : Plot a histogram
   * identify : Identify points on a plot with mouse clicks
   * plotmatrix : Plot an X vx Y scatter plot matrix
   * plotsym : Plot with symbols
   * qqnorm : Normal probability paper
   * qqplot : Create a QQ-plot
   * stairs : Stairstep graph
 * Logistic Regression
   * lodds : Log odds function
   * loddsinv : inverse of log odd
   * logitfit : Fit a logistic regression model
 * Miscellaneous
   * betainc : Incomplete beta function
   * betaln : Logarithm of beta function
   * corrcoef : Correlation coefficient
   * cov : Covariance matrix
   * ksdensity : Kernel smoothing density estimate
   * quantile : Empirical quantile
 * Polynomials
   * polyfit : Polynomial curve fitting
   * polyval : Polynomial evaluation
 * Random Numbers
   * rexpweib : Random numbers from the exponential or weibull distributions
 * Regression
   * cmpmod : Compare linear submodel versus larger one
   * lsfit : Fit a multiple regression normal model
   * lsselect : Select a predictor subset for regression
   * regres : Multiple linear regression
   * regresprint : Print linear regression
 * Resampling Techniques
   * stixbox_resamplingT : How to use extra-arguments in T.
   * ciboot : Bootstrap confidence intervals
   * covboot : Bootstrap estimate of the variance of a parameter estimate.
   * covjack : Jackknife estimate of the variance of a parameter estimate.
   * rboot : Simulate a bootstrap resample
   * stdboot : Bootstrap estimate of the parameter standard deviation.
   * stdjack : Jackknife estimate of the standard deviation of a parameter
 * Tests
   * ciquant : Nonparametric confidence interval for quantile
   * kstwo : Kolmogorov-Smirnov statistic from two samples
   * test1b : Bootstrap test that mean equals zero
   * test1n : Test that mean equals zero (Normal)
   * test1r : Test for median equals 0 using rank test
   * test2n : Tests two normal samples with equal variance
   * test2r : Test location equality of two samples using rank test
Files (2)
[450.96 kB]
Source code archive

[667.62 kB]
OS-independent binary for Scilab 5.4.x
Binary version
Automatically generated by the ATOMS compilation chain

News (0)
Comments (3)     Leave a comment 
Comment from Michael BAUDIN -- March 14, 2013, 10:12:54 PM    
Here is the changelog for v2.2. 

This is a major update, with new features, updates and many bug fixes.

stixbox (2.2)
    * Fixed bug #763: histo could not use the edges I want
    * Fixed bug #822: pairs had no example.
      Removed the pairs function: it did not work anyway. 
    * Created the plotmatrix function, compatible with Matlab.
      This is a replacement for the removed "pairs" function.
    * Removed function plotdens: it had no help page, no test, 
      and failed.
    * Created ksdensity function, compatible with Matlab.
      This is a replacement for the removed "plotdens" function.
    * For plotsym : clarified help page, implementation, 
      added examples, added unit tests.
    * Created new bubblechart, bubblematrix functions.
    * Created a "Polynomials" section.
      Added examples to polyfit.
      Added examples to polyval.
    * Changed implementation of polyval to better handle 
      ill-conditionned matrices, so that we can have a workaround 
      for bug :
      We use the linalg module, and the linalg_dgesv function.
    * Fix bug #188: polyfit and polyval are incompatible.
      Changed order of powers p in the argument of polyfit, 
      in order to 
       * make the call consistent with polyeval, 
       * make the function compatible with Matlab.
      This breaks the backward compatibility. 
      To update your code, please replace :
p = polyfit(x,y,n)


p = polyfit(x,y,n)

    * Fixed computation of confidence bound in polyval. 
      Added an example and a unit test of confidence interval.
    * polyfit : computed Vandermonde matrix from makematrix.
    * cov : changed implementation and help page. 
      Added examples in the help page.
      Added a unit test.
      cov is now compatible with Matlab.
    * Created new corrcoef function.
      corrcoef is compatible with Matlab.
    * Improved the stairs function to improve compatibility 
      with Matlab.
      stairs can plot several stairstep graphs in one statement.
      The stairs(y) calling sequence is available.
      Added a unit test.
      Added several examples in the help page.
    * Added the regres function, which performs multiple linear 
      This function is compatible with Matlab. 
      This function is similar to the "lm" function in R.
      The regres function was first available in the scidoe module. 
      The "regres" function name was chosen because "regress" 
      already exists in Scilab, but Scilab's "regress" is not Matlab-compatible,

      and has much less features.
    * Added the regresprint function, which prints the 
      analysis of variance table.
      This function is similar to the "anova" function in R.
    * Removed the linreg function. 
      This was done for the following reasons:
      * linreg did not work.
      * linreg was not compatible with Matlab.
      * linreg had no examples in the help page.
      * linreg had poorly documented input arguments.
      * linreg had no unit tests.
      * linreg did both graphics and computations, which was  
        a poor design choice.
      * linreg was univariate, while regress is multivariate.
    To update your legacy code, please replace:



B = scidoe_regress(Y,[ones(X),X]);

      But this is unlikely, since linreg did not work (at least, 
      the demo did not).
    * Updated the help page of identify.
      Added examples. 
      Added description of input arguments.
      Fixed a bug : the plotsymbol input argument was ignored.
    * histo : 
      Fixed a bug in the example where edges are set.
      Fixed a bug in the code where edges are set.
    * ciboot : 
      Updated the help page : added examples, parameters description.
      Updated the implementation to use distfun.
    * test1b : fixed implementation by replacing the 
      std function with Scilab's stdev.
    * Created a next "Resampling Techniques" section with:
      * ciboot
      * covboot
      * covjack
      * rboot
      * stdboot
      * stdjack
      Updated most help pages, with calling sequence, 
      parameter description, description and examples.
    * Fixed bug #1084:
      A new "Regression" help section was created.
    * Fixed a bug in covboot:
      The output values were wrong for "mean","median", since 
      Matlab returns columnwise results while Scilab 
      returns a scalar.
      The estimate in this case was completely wrong. 
      For "std", the function did not work, as this function 
      does not exist in Scilab.
    * Fixed bug #1083:
      The Resampling Technique functions had no unit tests.
    * Fixed bug #1070
      The Resampling Technique functions do not check input args.
    * Fixed bug #1069:
      The Resampling Technique functions do not manage 
      extra-arguments properly.
    * Fixed bug #1071
      rboot had no example
    * Fixed the test1b function : the confidence interval for the 
      standard deviation was wrong. 
      This is because the stdev function in Scilab returns a scalar, 
      while Matlab's std returns a row vector.
    * Added help page and unit test for test1b.
    * Added help page and unit test for test1n.
      Fixed implementation : pt, qt and qchisq did not 
      exist anymore.
      Fixed bug #1086: Execution of stixbox 2.1-1 test1n function failed.
    * Added help page and unit test for test2n.
      Fixed implementation : var, pt, qt and qchisq did not 
      exist anymore.
    * Added help page and unit test for test1r.
    * Added help page and unit test for test2r.
Comment from Vanesa Lo Iacono Ferreira -- January 15, 2014, 01:24:07 PM    
Excellent improvements! Thanks. 
Comment from David Chèze -- October 6, 2014, 04:54:25 PM    
Thanks for this great module.
Is there on going development to introduce robust boxplot plotting function in this 
module ? I've tested here and there some alternative that failed most of the time.

Leave a comment
You must register and log in before leaving a comment.
Login with GitLab
Email notifications
Send me email when this toolbox has changes, new files or a new release.
You must register and log in before setting up notifications.