Projects / CCruncher

CCruncher

CCruncher is a project for the simulation of large portfolios of SME loans where the unique risk is the default risk. The method used to determine the distribution of losses in the portfolio is the Monte Carlo algorithm, because it allows you to consider multiple variables, such as the date and amount of each payment. The obligors' default times are simulated using a copula with given survival rates and correlations.

Tags
Licenses
Operating Systems
Implementation

RSS Recent releases

Release Notes: This release adds support for macros (useful for sensitivity analysis), solves the precision problem in the survival functions, modifies the t-Student CDF speedup algorithm, and updates the technical document.

Release Notes: This release adds support for stochastic exposure, improves performance significantly, and solves minor bugs.

Release Notes: This version adds support for stochastic recoveries, adds support for simple and continuous interest rates, and considere exposure instead of cashflow events. It also improves the report and solves minor bugs.

Release Notes: This version improves t-Student copula fit, replaces MPI support by multi-threading, and improves performance. Also, some minor changes in input file format and command line arguments have been made.

Release Notes: This release adds transition matrix regularization and the computation of the condition number for the Cholesky matrix. Performance has been improved in the file parsing, the copula generation, and the portfolio aggregation. The input file format was also changed to make it more readable.

Screenshot

Project Spotlight

BitNami LAMPStack

BitNami LAMPStack Native Installer is an easy-to-install environment to develop and deploy PHP applications. It includes pre-configured, ready-to-run versions of Apache, MySQL, PHP, and phpMyAdmin.

Screenshot

Project Spotlight

RPL/2

A programming language for computations.