CAT - Credit Analysis Tool
IBM PureFlex System
|Provided by: Quantitative Risk Research SL|
A complete environment for the analysis and optimization of credit portfolios. It is able to obtain the optimal portfolio based on the VaR attribution of each position of the portfolio.
QRR CAT: A credit analysis tool Portfolio Analysis and optimization
Placed in a Merton-Vasiseck multifactor framework, the analysis performs the calculation of the credit risk measures at different aggregation levels (global, by geography, by sector and individually), taking into account both the effects of diversification as the idiosyncratic risks of individual positions. Capital attribution is also included.
Two approaches for the evaluation of the risk metrics are included:
- An approximate analytic approach based on the formulation given by Pykhtin for the estimation of the VaR and the CVaR, that includes the multifactor and portfolio granularity effects. This method can be up to 100x faster than Monte Carlo based approaches.
- A Monte Carlo approach based on the simulation of every risk factor in the model. It allows using richer models, such as using a stochastic LGD or incorporating large counterparties with high exposure.
Portfolio can be optimized by means of the maximization of economic profit subject to given restrictions applied at different levels of aggregation (individually or by geography).
Reports with the analysis and optimization results can be exported to files in the Excel format.
Java J2EE Web application.
Optimal numeric performance by using C/C++ optimization libraries and CUDA (GPGPU) for Monte Carlo.
- Calculation of the portfolio risk measures (economic capital, expected, loss, ...) at different levels of aggregation.
- Analysis based on Merton-Vasiseck multifactor model.
- Two approaches: an approximate analytic approach based on the formulation given by Pykhtin for the attainment of the VaR and a Monte Carlo model based on the simulation of every risk factor in the model.
- Optimization of the portfolio based on the economic profit subject to given restrictions.
- Use of efficient C/C++ and CUDA (GPGPU) libraries for optimal numerical computations.
- Data load is performed from spreadsheet files. Portfolio parameters are editable using the web interface.
Name: Santiago Carrillo
Phone: +34 91 806 48 42