Upload
hoangquynh
View
221
Download
5
Embed Size (px)
Citation preview
proven risk intelligence for lenders and investors 800.333.4510 interthinx.com [email protected]
©2012. All rights reserved. Interthinx is a registered trademark of Verisk Analytics. Mortgage Loan Assessment, LookAhead, and TrueCapital are trademarks of Interthinx, Inc., FraudGUARD, ValueGUARD, Clear Value, PredProtect are registered trademarks of Interthinx, Inc..
ProvenMortgage Fraud Mitigation & Verification Services
FraudGUARD®
SafeCheck®
SSN Direct4506-T
ExpertCollateral Risk & Valuation Solutions ValueGUARD®
Interthinx Review Appraisal ServicesClear Value® AVMiAVM Plus
ReliableRegulatory Compliance
PredProtect®
TPRRegs DataCivil/Criminal Background Checks
AuthenticLoan ReviewSolutions
Repurchase ReviewsMI RescissionQuality ControlCollateral Value ReviewTPO ReviewCompliance ReviewFraud Investigation
VisionaryPredictive Analytics
MBS & ABS Forecasting Mortgage Risk Model™ LookAhead™
Portfolio Forecasting TrueCapital™
Capital Planning
Interthinx End-to-End Risk Mitigation As the largest aggregator of mortgage application data and the most comprehensive proprietary database, Interthinx provides proven end-to-end risk mitigation tools and a leading team of experts in mortgage fraud, collateral risk, regulatory compliance, retail portfolio forecasting and audit services.
PredictiveAnalytics
Dual-time DynamicsProven forecasting technology laying
the foundation for Interthinx’s
Predictive Analytics solutions
The unique Dual-time Dynamics (DtD) technology behind Interthinx’s Predictive Analytics solutions and services,
including LookAhead™, decomposes and quantifies each of the
key drivers of portfolio performance: lifecycle, environment,
and credit quality.
DtD technology’s decomposition is the foundation for:
• Gaininginsightintowhatcontributed to past portfolio performance.
• Customizationofforecaststo account for future origination and economic scenarios.
• Confidenceandtransparency in forecasting models.
Pro
prie
tary
800.333.4510 [email protected] more information about Dual-time Dynamics technology or other products, please contact: interthinx.com
PredictiveAnalytics
THEDUAL-TIMEDYNAMICSAPPROACH
Based on a patented nonlinear algorithm, Dual-time Dynamics automatically measures factors driving portfolio performance from historical performance data. The lifecycle, environment, and vintage quality components measured by Dual-time Dynamics provide a unique view into the factors driving portfolio performance and serve as individual controls on scenarios that will drive future performance. Traditional portfolio models assume that a predetermined set of variables drives portfolio performance. These models are biased towards the selected model variables and the performance period of the data used to train the model. Dual-time dynamics makes no assumptions about which factors drive portfolio performance. Instead, it measures performance along the dimensions of age, time, and origination date. Dynamics such as life cycles and seasonality tend to be stable over time, enabling users to focus on marketing and economic scenarios to drive forecasts.
Separating portfolio drivers into lifecycle, vintage quality, seasonality, policy changes, and economics provides unprecedented flexibility in using scenarios to drive portfolio forecasts. Users are free to choose economic indicators for forecasting and specify originations plans. Scenario components are automatically combined to produce forecasts. Users can run forecasts against multiple economic scenarios to stress test portfolios and Dual-time Dynamics can quantify the amount that each scenario component contributes to the forecast.
The factors driving portfolio performance in the future are different than the factors that drove performance in the past. Measuring and forecasting changing conditions is paramount in effective portfolio management.
Key Advantages of Dual-time Dynamics
USMORTGAGECRISIS
In 2005 an analysis done using DtD identified extreme risk in the US mortgage market based on deteriorating quality of originations and natural loan maturation. Losses were projected to exceed industry expectations even under the assumption that the relatively healthy economic conditions would continue. This analysis highlighted the need for additional capital reserves prior to the onset of deterioration in the overall mortgage market in 2007.
CRISISTESTEDAROUND THEGLOBE
DtD technology is used in more than 25 countries on hundreds of portfolios. Its forecasts remained robust through the 2001 Global Recession, the 2003 Hong Kong SARS Recession, the US Great Recession of 2009 and the 2009 Global Financial Crisis. Additionaly, it has been back-tested on the Asian Economic Crisis of 1997.
PEERREVIEWEDANDVALIDATED
The patented technology for Dual-time Dynamics has been published in eight peer-reviewed academic articles, presented at numerous academic conferences and demonstrated to key regulatory bodies around the world. Many of these publications and presentations have focused on validation of every aspect of Interthinx’s Predictive Analytics solutions and the standards by which other models should be evaluated.
ROBUSTANDINTEGRATED
Even on short, thin, or noisy data sets the robust model estimation approach used by Dual-time Dynamics technology automatically creates forecasts. DtD technology’s model components have become industry-standard terms and the modeling framework provides a common approach across all retail loan types and geographies, giving management the ability to leverage common reports and terminology.
EXOGENOUS
Date
Relative
Change
VINTAGEDECOMPOSITION
Months-on-Books
AccountFlowT
hro
ugh
60-8
9D
PDR
ate
MATURATION
ANNUALVINTAGES
QUALITY
Vintage
Relative
Change
DtDwillmeasurethefollowingcomponentsfrom vintage performance date:
LIFECYCLE
Each product type (5-year auto loan or 30-year fixed mortgage for example) has its own unique lifecycle describing when delinquencies increase, peak and fall back down. The lifecycle effect per product type is generally very stable when environmental and loan quality impacts are correctly factored into the calculation.
CREDITQUALITY
The quality of each vintage will vary based on originations criteria and market conditions. Because the quality measure does not depend on scores, it can detect changes in these conditions.
ENVIRONMENT
The exogenous or environment component includes internal management actions, external regulatory changes, seasonality and macroeconomic fluctuations that affect portfolio performance. With the decomposition of the exogenous curve and special correlation to the macroeconomic factors, it is possible to stress test forecasts based on multiple economic scenarios.
PredictiveAnalytics800.333.4510 [email protected] more information about Dual-time Dynamics technology or other products, please contact: interthinx.com