Statistics and Data Analysis

Integration Problems

The goal of this toolbox is to provide integration test problems in order to test integration algorithms. This allows testing integration algorithms on a larger base of functions. Developing these functions is error-prone. The use of this toolbox saves time and ensures accuracy of the results.

Moreover, testing an algorithm involves choosing a set of test functions. But all functions are not equally difficult to integrate; while some are simple, others might be particularly difficult, leading to poor results. Comparing two algorithms, each on a different set of functions will not lead to an accurate view of the algorithm. This module enables the comparison of the algorithms more precisely.


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

Stixbox includes Probability Distribution Functions, Probability Cumulative Distribution Functions, Inverse Cumulative Distribution Functions, Random Numbers, Graphics and Logistic Regression.


This module includes scilab functions for polynomial  regression and linear regression over multiple independent variables.

Microwave Toolbox 

This toolbox enables efficient data analysis of RF and microwave data,  such as S2P file read/write and display, stability circles, de-embedding. This toolbox can be used to manage measurement data. The microwave toolbox includes several functions for S2P touchstones and MES/MPS files. The toolbox can be used to display S parameters, IV-networks and to perform de-embedding  with just one command.


Metanet is a Scilab toolbox to do graph and network computations.  A number of algorithms solving classical graph problems and minimal cost flow network are provided. The toolbox has functions to add an edge or an arc between two nodes, associate new data fields to the edges data structure of a graph, add disconnected nodes to a graph, plot graphs, replace a group of nodes within one node, perform union of two graphs.


This toolbox is used to do sensitivity analysis. This is the analysis of the uncertainty in the outputs of a given model, depending on the uncertainty in its inputs.

The analysis is based on chaos polynomials, the orthogonal polynomials that are used as an approximation of the original model. The sensitivity indices are computed from the coefficient of the chaos polynomial.

Artificial Neural Network(ANN) Toolbox

The toolbox includes the following features:

- Layered feedforward networks are supported directly at the moment [Use the hooks provided to represent other types of networks]

- Unlimited number of layers

- Unlimited number of neurons in each layer

- User defined activation function

- User defined error function

Scilab Wavelet Toolbox

Wavelet is a powerful Signal processing tool. The Scilab Wavelet Toolbox has the following feature

  • Discrete Fast Wavelet Transform, daubechies wavelets
  • 1-D single level signal decomposition and reconstruction
  • 1-D multi-level signal decomposition and reconstruction
  • 2-D single level image decomposition and reconstruction
  • 2-D multi-level image decomposition and reconstruction
  • 3-D single level signal decomposition and reconstruction
  • stationary wavelet transform
  • continuous wavelet transform
  • dualtree real and complex wavelet transform

Low Discrepancy Sequences

The goal of this toolbox is to provide a collection of low discrepancy sequences. These random numbers are designed to be used in a Monte-Carlo simulation. For example, low discrepancy sequences provide a higher convergence rate to the Monte-Carlo method when used in numerical integration. The toolbox takes into account the dimension of the problem, i.e. generate vectors with arbitrary size.

The toolbox has the following features

  • manage arbitrary number of dimensions,
  • skips a given number of elements in the sequence,
  • leaps (i.e. ignores) a given number of elements from call to call,
  • fast sequences based on compiled source code,
  • suggest optimal settings to use the best of the sequences.


GROCER is used in econometric Studies. The toolbox includes ols, instrumental variables, VAR, specification and unit root tests, limited dependent variables methods, Johansen cointegration tests and VECM, GMM, statistical filters, basic methods on panel data, ARMA and VARMA estimations, GARCH …

GROCER also provides less standard methods: a pc-gets like function implementing in an automatic manner the “general-to-specific” methodology, various Markov-switching estimations, Bayesian Model Averaging estimation, dynamic contributions.

GROCER comes with 2 new types of data useful to an econometrician: time series and matrices of time series