CreditCruncher - FAQ

1. What try to solve CreditCruncher ?

CreditCruncher is a program that uses the Monte Carlo method to compute the credit risk of large portfolios in which assets are mortgages, loans, bonds, endorsements, or the like (all of them fixed income with a policy buy/sell and hold). The default time is simulated using a gaussian copula, taking into account the transition matrix (or survival functions) and sectorial correlation matrix defined by the user.

2. What is the project maturity ?

CreditCruncher-1.1 is a stable version. If you have any bug or suggestion to report, please send a mail to ccruncher team or post it to SourceForge.

3. Can I found docs where the CreditCruncher algorithm is explained?

Yes. The document is named CreditCruncher - Technical Document and you can find a link here. This document is included in all the distributions under the directory docs/.

4. Windows platform is suported ?

Yes, but the main stream is in unix environment. If you desire to compile ccruncher under Windows, read the file docs/README. The repository contains the project files for MSVC7 compiler (*.sln, *.vcproj).

5. Why input files can be gziped (*.gz) but not ziped (*.zip) ?

zip algorithm is covered by patents, but gzip is not; furthermore gzip is open source. You can download a gzip tool for windows from 7-zip. Usually, unix environments have gzip/gunzip commands. You can read more at gzip home.

6. CreditCruncher is fast ?

Yes. CreditCruncher is designed to be very fast. It allows you to evaluate the credit risk of your whole portfolio (there can be more than 50.000 borrowers) each day. If ccruncher sunk your computer, you can execute it on a cluster enabling the MPI compilation option.