You are here: home | Stixbox | 2.1

Stixbox

Statistics toolbox

(937 downloads for this version - 64094 downloads for all versions)

Details

Version

2.1
A more recent valid version with binaries for Scilab 5.4 exists: 2.4

Authors

Anders Holtsberg

Paris Sud University

Maurice Goursat

Jean-Philippe Chancelier

Serge Steer

Allan Cornet

Michael Baudin

Owner Organization

INRIA,DIGITEO, Université Paris Sud, and others

Maintainers

Allan CORNET

Michael BAUDIN

Categories

Data Analysis - Statistics

Data Analysis - Statistics - Statistics

Website

http://forge.scilab.org/index.php/p/stixbox/

License

CeCILL

Dependencies

Specfun

Apifun

Helptbx

Distfun

Creation Date

February 22, 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")

Report a bug

GitLab's issues for stixbox

Description

            Purpose
-------

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

Features
--------

 * Cumulated Distribution Functions
   * pks : Kolmogorov Smirnov distribution function 
 * Random Numbers
   * rexpweib : Random numbers from the exponential or weibull distributions 
 * Datasets
   * getdata : Famous datasets 
 * Graphics
   * histo : plot a histogram 
   * identify : Identify points on a plot by clicking with the mouse 
   * linreg : Linear or polynomial regression 
   * pairs : Pairwise scatter plots 
   * plotdens : Draw a nonparametric density estimate 
   * plotsym : Plot with symbols 
   * qqnorm : Normal probability paper 
   * qqplot : Plot empirical quantile vs empirical quantile 
   * stairs : Plots the Empirical Cumulated Distribution.
 * 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 
   * ciboot : Various bootstrap confidence intervals 
   * cov : Covariance matrix 
   * covboot : Bootstrap estimate of the variance of a parameter estimate 
   * covjack : Jackknife estimate of the variance of a parameter estimate 
   * polyfit : Polynomial curve fitting 
   * polyval : Polynomial evaluation 
   * quantile : Empirical quantile (percentile). 
   * rboot : Simulate a bootstrap resample from a sample 
   * stdboot : Bootstrap estimate of the parameter standard deviation 
   * stdjack : Jackknife estimate of the standard deviation of a parameter
estimate 
 * Tests, confidence intervals and model estimation
   * ciquant : Nonparametric confidence interval for quantile 
   * cmpmod : Compare linear submodel versus larger one 
   * kstwo : Kolmogorov-Smirnov statistic from two samples 
   * lsfit : Fit a multiple regression normal model 
   * lsselect : Select a predictor subset for regression 
   * test1b : Bootstrap t test and confidence interval for the mean 
   * test1n : Tests and confidence intervals based on a normal sample 
   * test1r : Test for median equals 0 using rank test 
   * test2n : Tests and confidence intervals based on two normal samples with
common variance

Files (2)

[127.08 kB]

stixbox-2.1-1-src.zip

Source code archive

Upload date : 2013-02-22 23:19:12
MD5         : 303272c224b1ee341331a038499a4341
SHA1        : 4f55e9fdb50a15c6fd2739e4903bff79e156af58
Downloads   : 219
 File list

[257.56 kB]

stixbox_2.1-1.bin.zip

OS-independent binary for Scilab 5.4.x

Binary version
Automatically generated by the ATOMS compilation chain

Upload date : 2013-02-25 09:25:08
MD5         : b84b9bb116a299784357d715770acc4b
SHA1        : b46466fe030766910eead0325e71788876e07608
Downloads   : 718
 File list

News (0)

Comments (3) ↓ Leave a comment

Comment from Michael BAUDIN

-- February 22, 2013, 11:19:57 PM

Here is the changelog.

stixbox (2.1)
    * Fixed bug #1025 : qqplot had no example
      http://forge.scilab.org/index.php/p/stixbox/issues/1025/
      Improved the help page of qqplot:
      added several examples.
    * Fixed bug #1024: The example in histo does not work.
      http://forge.scilab.org/index.php/p/stixbox/issues/1024/
    * Removed bincoef (obsolete).
To update your code, please replace

bincoef(n,N)

with 

specfun_nchoosek(N,n)

    * Removed gammainc (obsolete).
To update your code, please replace

y=gammainc(x,a)

with 

y=distfun_gammainc(x,a)

Comment from Chevillard eric -- March 12, 2013, 04:55:13 PM

bonjour,
La version stixbox 2.1-1 n'a plus comme macro la fonction qt (et d'autres ?).
Résultat linreg ne fonctionne plus avec cette version.

Je suis revenu à la version 1.2.5 qui fonctionne pour linreg. Merci de corriger ce
défaut.

Comment from Michael BAUDIN

-- March 18, 2013, 09:22:50 PM

Hi,

Thank you for your interest on this module.

The v2.1 of Stixbox has, indeed, no function qt anymore. This is since the v2.0 and the 
reasons for this change are explained in the changelog of the v2.0, available at:

http://atoms.scilab.org/toolboxes/stixbox/2.0

This set of distributions was of poor quality in Stixbox. To convince yourself, please 
read the bug report [1], which shows that the PDF for the Normal distribution was wrong 
right from the start. There were several other important bug reports on this topic in 
Stixbox, so that maintaining these functions had no technical sense. Distfun is much 
better on this topic and your scripts (and numerical results) will improve by using 
Distfun. 

On the linreg function, I removed this function in the v2.2. The reasons for this change 
are explained in the changelog of the v2.2:

http://atoms.scilab.org/toolboxes/stixbox/2.2

The v2.2 has a new "regres" function, which is Matlab-compatible and replaces
"linreg".

You can technically keep on using the v1.2.5. But I strongly suggest to update your 
module: the old version has more than 26 bug identified and fixed (and many unidentified),

while the version 2.2 has unit tests to prove that the new function really work on real 
data.

Please enter your bug reports at:

http://forge.scilab.org/index.php/p/stixbox/issues/

Best regards,

Michaël

[1] http://forge.scilab.org/index.php/p/stixbox/issues/164/