Course Agenda


Model Risk Management - Toronto

Course Agenda

Day One


Registration and refreshments


Model Risk Management & Governance  

  • Definitions of models and model risk 

  • Model types and model approaches 

  • Sources and key drivers of model risk 

  • Evolution of model risk management 

  • OSFI guidelines and other regulations on model risk management

  • Model governance framework including three lines of defence 

  • Coverage of model workflows including model interconnectivity 

Speaker: Yaping Jiang, Managing Director, JW Matrix


Morning break


How to Build a Model Risk Management Framework 

  • Development, quantification, integration, implementation 

  • Setting risk appetite, policy & standards for model risk 

  • Model inventory process 

  • Model lifecycle management (development, validation, implementation, use, periodic review) 

  • Estimating capacity for risk 

Speaker: Olga Streltchenko, Director, Export Development Canada




Model Validation & Performance Analysis 

  • What is validation? 

  • Improving the models 

  • Validation tools 

  • Performance analysis review 

  • How to quantify model limitations 

  • Vendor and third-party model validation 


Afternoon Break


Model Risk Management for Pricing Models

  • Best approach to pricing models 

  • Products in balance sheet 

  • Market of products vs. pricing and hedging 

  • Source of valuation adjustments in pricing

  • Identification and mitigation of model and input risk 

  • Establishing pricing and validation framework


End of day one

Day Two - Thursday, 26 September




Model Risk Management of Non-Pricing Models 

  • Retail models (credit scoping/marketing) 

  • IFRS 9 overview & progress since implementation 

  • The similarities and difference of CECL compared to IFRS 9 and other regulatory credit models 

  • Sources of model risk in IFRS 9 models 

  • What new strategies need to be put in place for testing IFRS 9 models and assumptions? 

Speaker: Grigoris Karakoulas, President and Founder, InfoAgora Inc.


Morning break


Utilizing Machine Learning for Model Validation and Monitoring of Valuation models

  • Why banks need larger validation throughput and how to use AI to speedup 

  • Measuring data quality with ML 

  • Building AI challenger models for model risk uncertainty measurement 

  • Generating test scenarios with ML 

  • Solving PDE’s with deep reinforcement learning 

  • Validation with AI of market data generation algorithms (IR curve building, volatility surface construction) 

  • Monitoring valuation models with ML (PnL & XVA) 




A Deep Dive into Validation of Natural Language Processing Models

  • Statistical foundations of NLP and implications for model risk governance 

  • Use case: word embeddings and recurrent neural networks 

  • Validation techniques for models with text input

  • Responsible AI: fairness and interpretability 

  • Applications of recent advances in Machine Learning for NLP validation 

Speakers: Greg Kirczenow, Senior Director - Model Risk Management, RBC

                  Ali Fathi, Senior Manager AI Research, RBC 


Afternoon Break


Model Risk into the Future 

  • Applying models to new challenges 

  • Data challenges 

  • Automation vs. human judgement 

  • Big data and advanced analytics 

  • Treatment and governance of near-models and non-models 

  • Future of regulation; possible futures

  • Further evolution of models 


End of Course