CCruncher is a project for quantifying portfolio credit risk using the copula approach. It is a framework consisting of two elements: a technical document that explains the theory, and a software program that implements it. CCruncher evaluates the portfolio credit risk by sampling the portfolio loss distribution and computing the Expected Loss (EL), Value at Risk (VaR), and Expected Shortfall (ES) statistics. The portfolio losses are obtained simulating the default times of obligors and simulating the EADs and LGDs of their assets.
Armadillo is a C++ linear algebra library (matrix maths) aiming towards a good balance between speed and ease of use. The API is deliberately similar to Matlab's. Integer, floating point, and complex numbers are supported, as well as a subset of trigonometric and statistics functions. Various matrix decompositions are provided through optional integration with LAPACK and ATLAS numerics libraries. A delayed evaluation approach, based on template meta-programming, is used (during compile time) to combine several operations into one and reduce or eliminate the need for temporaries.
Lynkeos processes astronomical webcam images. By stacking the best images, the signal to noise ratio is increased, and details lost in the noise of individual images become visible in the resulting image. This software accepts, as input, QuickTime, AVI, and MPEG sequences, or still images, in any image format supported by its active plugins (TIFF, FITS) and Cocoa. It generates a 16-bit RGB TIFF or a monochrome FITS image as output, to be further processed with some all-purpose image processing application.
Social Networks Visualizer (SocNetV) is a flexible and user-friendly tool for the analysis and visualization of Social Networks. It lets you construct mathematical graphs with a few clicks on a virtual canvas, load networks of various formats (GraphViz, GraphML, Adjacency, Pajek, UCINET, etc), or create a network by crawling all links in a Web page. The application can compute basic network properties, such as density, diameter, and distances (shortest path lengths), as well as more advanced structural statistics, such as node and network centralities (i.e. closeness, betweenness, graph), clustering coefficient, etc.
iNA is a computational tool for quantitative analysis of fluctuations in biochemical reaction networks. Such fluctuations, also known as intrinsic noise, arise due to the stochastic nature of chemical reactions and cannot be ignored for when some molecules are present only in very low copy numbers as is the case in living cells. The SBML-based software computes statistical measures such as means and standard deviations of concentrations within a given accuracy using the analytical system size expansion. The result of iNA’s analysis can be tested against the computationally much more expensive stochastic simulation algorithm.
SHOGUN is a machine learning toolbox whose focus is on large scale kernel methods and especially on Support Vector Machines (SVM). It provides a generic SVM object interfacing to several different SVM implementations, all making use of the same underlying, efficient kernel implementations. Apart from SVMs and regression, SHOGUN also features a number of linear methods like Linear Discriminant Analysis (LDA), Linear Programming Machine (LPM), (Kernel) Perceptrons, and algorithms to train hidden Markov models. SHOGUN can be used from within C++, Matlab, R, Octave, and Python.