Risk Management Toolbox
Develop risk models and perform risk simulation
Risk Management Toolbox™ provides functions for mathematical modeling and simulation of credit and market risk. You can model probabilities of default, create credit scorecards, perform credit portfolio analysis, and backtest models to assess potential for financial loss. The toolbox lets you assess corporate and consumer credit risk as well as market risk. It includes an app for automatic and manual binning of variables for credit scorecards. It also includes simulation tools to analyze credit portfolio risk and backtesting tools to evaluate value-at-risk (VaR) and expected shortfall (ES). You can model lifetime probability of default (PD) to estimate loss reserves for lifetime credit analysis.
Get Started:
Stress Testing
Perform stress testing and sensitivity analysis on financial portfolios.
Lifetime Expected Credit Loss Modeling
Estimate lifetime expected credit losses in compliance with risk regulations such as CECL and IFRS 9.
Lifetime probability of default for a stress test.
Calculating Regulatory Capital
Calculate capital requirements and value-at-risk with the asymptotic single risk factor (ASRF) model.
Regulatory capital by asset class.
Credit Scorecards Modeling
Identify the variables in your data sets that have the best predictive power using tools for predictor screening. Once you’ve identified important variables, use the Binning Explorer app to develop credit scorecards by applying auto-binning algorithms or interactively adjusting edges, merging bins, and splitting bins. You can also fit a logistic model, obtain points and score, and calculate the probability of default. Once developed, deploy a lightweight version of the model using compact credit scorecard.
Binning Explorer app for credit scorecard modeling.
Credit Risk Simulation
Perform copula simulations based on probability of default or credit rating migration to analyze the risk of credit portfolios. Simulation throughput can be increased through parallel computing using Parallel Computing Toolbox.
Portfolio losses based on copula simulations.
Risk Parameters Estimation
Estimate probability of default (PD) using various methods, including structural models, reduced-from models, historical credit rating migration, and other statistical approaches. Use the lifetime probability of default (PD) models to estimate the loss reserves based on a lifetime analysis conditioned on macroeconomic scenarios. Additionally, you can use Risk Management Toolbox to calculate concentration risk indices.
Lorenz curve for representing the distribution of risk exposure.
Value-at-Risk Backtesting
Risk Management Toolbox VaR backtesting models include traffic light, binomial, Kupiec's, Christoffersen's, and Haas' tests.
Results from multiple VaR backtesting models.
Expected Shortfall Backtesting
Backtesting models for expected shortfall (ES) include conditional test, unconditional test, quantile test and minimally biased test by Acerbi and Szekely, as well as conditional and unconditional tests by Du and Escanciano
Historical VaR and ES plot.
Claims Estimation
Use the development triangle along with other estimation techniques such as chain ladder, expected claims, and Bornhuetter-Fergurson to estimate unpaid claims.
Development of the reported claims
Product Resources:
Computational Finance Suite
The MATLAB Computational Finance Suite is a set of 12 essential products that enables you to develop quantitative applications for risk management, investment management, econometrics, pricing and valuation, insurance, and algorithmic trading.
