Managing risks and improving profits in a world of uncertainties
Retail lending has changed dramatically in the last couple of decades. Machine Learning and AI bring new capabilities, regulations like IFRS 9 require model enhancements, and all models receive much greater audit scrutiny. Reliance on weak, out-dated models can create portfolio disasters. Managing a retail loan portfolio successfully depends on properly developing, validating, deploying, and integrating a wide range of models and new technologies.
This course is for analysts, managers, and credit professionals who are involved in model development, validation, or model risk management. Although the course covers the development and use of sophisticated analytical techniques, it is intended for portfolio managers, financial analysts, credit policy professionals, and statistical analysts. Anyone with decisionmaking responsibility in this field will benefit. Discussions will minimize mathematical derivations, instead emphasizing intuitive understanding, use of available algorithms, and best practices in application and implementation.
- Learn the fundamentals of retail lending analytics.
- Gain an understanding for when these methods can be used reliably, when they fail, and how to use new methods to succeed.
- Obtain a deep understanding of the drivers of credit risk, and how various models capture some, or all, of these drivers.
- Learn how to leverage the latest techniques in machine learning with an eye to successful validation and regulatory examination.
- Acquire best practices for building models for IFRS 9, stress testing,
Le Meridien, 2 Jalan Stesen Sentral, Kuala Lumpur Sentral, KL 50470 Malaysia
Joseph L. Breeden Ph.D., Chief Executive Officer, Deep Future Analytics
LLC, www.deepfutureanalytics.com
Dr. Breeden has 25 years of experience in financial services and is a recognized leader in the industry. At DFA, he is directly involved in research into new modeling techniques and products. He co-founded Strategic Analytics in 1999, where he led the design of advanced analytic solutions including the invention of Dual-time Dynamics.
Dr. Breeden has created models through the 1995 Mexican peso crisis, the 1997 Asian economic crisis, the 2001 global recession, the 2003 Hong Kong SARS recession, and the 2007–2009 U.S. mortgage crisis and global financial crisis. These crises have provided Dr. Breeden with a rare perspective on crisis management and the analytics needs of executives for strategic decision-making. Dr. Breeden’s own models were successful throughout the U.S. mortgage crisis and warned of problems as early as the beginning of 2006.
His book Reinventing Retail Lending Analytics: Forecasting, Stress Testing, Capital, and Scoring for a World of Crises was published by Riskbooks in 2010. He currently serves as associate editor for the Journal of Risk Model Validation and the Journal of Risk Model Validation.
You should be comfortable with the basics of retail lending in at least one area and have some familiarity with building models in banking and have a basic understanding of credit scoring.
For questions about this course, please contact Dr. Joseph Breeden, breeden@deepfutureanalytics.com
Day 2 (9:00 a.m. – 5:00 p.m.)
Model Types Overview
Collective vs individual assessment
Comparison of strengths and weaknesses
Vintage Modeling
Age-Period-Cohort Decomposition
Macroeconomic Sensitivities
Quantifying Adverse Selection
Account-level Modeling
Logistic Regression
Survival Models
Multihorizon Survival Models
Model Risk Management
Conceptual Soundness
Effectiveness
Sensitivity
IFRS 9
Stages of impairment
12-month vs lifetime expected credit losses
Techniques and tips
Issues by asset class
Stress Testing
Climate Change
Pandemics
Economic Cycles
Capital
Regulatory Capital (Basel II)
Economic Capital
Pricing and Profitability
Paydown / payoff / attrition models
Interest and fees
Recoveries
Optimizing yield versus volume
Machine Learning Best Practices
Which methods work where, and why
ML failure risks
ML compliance risks
RM 2350 for the course including meals and snacks
Full refunds will be available on all cancellations mailed or faxed to the registrar up to 15 working days prior
to the start of the event. Registrations cancelled 6–14 working days prior to the event are subject to a
service fee equal to 50% of the registration fee. Registrants who cancel reservations 5 or fewer working days
prior to the event will forfeit the entire fee. Registrants failing to attend the event—no-shows—will not be
eligible for refunds.