Non-Agency MBS

Talk to us to find out which product best meets your needs
TOOLS TO MANAGE THE RISK IN YOUR NON-AGENCY MBS

At AD&Co, our tools use both traditional, as well as robust alternatives to evaluating risk in non-Agency MBS. Our proprietary models help value fixed or adjustable loans, as well as collateralized or pass-through loans to provide the confidence you need in making decisions to either retain or shed mortgage credit risk. Models can also be tuned to meet the specific needs of your portfolio.
 

Our products can help you:

  • Manage interest rate and credit risk in your mortgage portfolio

  • See how the forecasts change based on shifting economic and market conditions such as home prices and interest rates

Tools you can use

LoanDynamics logoNon-Agency LoanDynamics is our proprietary behavioral model that helps to manage asset liability, forecast credit loss, determine MSR valuation, allow for pipeline hedging, and support both secondary marketing and portfolio management.

LoanKinetics logoLoanKinetics is our proprietary residential whole loan analytical application that allows users to evaluate legacy and new residential loans for credit performance, valuation, loan loss analysis/attribution and stress testing.

RiskProfiler logoRiskProfiler is our comprehensive valuation solution that integrates our LoanDynamics and RiskValDynamics models into a single, flexible platform. It offers built-in reporting and parallel and distributed (cluster) computing, as well as a database in which to store positions and past performance.

Insights that can help you
Contact our Client Support Team for access
The S-Curve: AD&Co's blog

Subscribe to our blog for news that affects you, including product updates, industry events and our latest research.

Mortgage Analysis and Reporting System (MARS)

MARS is our web-based reporting system that provides an objective view of actual vs. forecasted results on an on-demand basis. Users can specify the product by vintage and analysis period, coupon bucket or as-of date to generate user-defined displays of how the model is performing versus actual speeds.