Model validation
Clearing houses warn Esma margin rules will stifle innovation
Changes in model confidence levels could still trip supervisory threshold even after relaxation in final RTS
For tomorrow’s quants, Python is essential; AI isn’t
Proportion of PhDs in quant teams is sliding, as employers focus on all-round skills
An aggregated metrics framework for multicriteria model validation using rolling origin evaluation
The authors apply the rolling origin evaluation framework to model validation in multicriteria settings, where performance must be assessed through various scenarios or forecast targets.
Probabilistic classification with discriminative and generative models: credit-scoring application
The author investigates how probabilistic classification can be used to enhance credit-scoring accuracy, offering a robust means for assessing model performance under various reliability criteria
A comprehensive explainable approach for imbalanced financial distress prediction
The authors suggest an explainable machine learning method for imbalanced financial distress prediction which uses extreme gradient boosting.
Speedy onboarding: the push for faster model approvals
Europe’s banking watchdog is planning to streamline how it authorises credit model updates. Not a moment too soon, say bankers
Generative AI brings testing times for modellers
Flagstar’s lead model validator offers some tips for safely integrating LLMs into risk models
Brain drain at OCC raises concerns about US model supervision
Quant team cull will reduce capacity to validate bank models, but that could be part of the plan
Model risk quantification for machine learning models in credit risk
This paper analyses bank-specific model risk measurement methods with a focus on implemented model risk rating solutions for MLMs and discusses challenges faced by the validation function.
AI shows cognitive bias just like humans, tests show
Risk Live: New form of op risk may be “especially dangerous” for model validators, quant says
A three-stage fusion model for predicting financial distress considering semantic and sentiment information
The authors apply sentiment analysis to management discussion and analysis texts to aid the prediction of financial distress with an innovative three-phase fusion model.
Emir rule delay leaves Simm paperwork gathering dust
Mid-year refresh triggers Emir 3.0 authorisation process despite unfinished regulatory standards
JP Morgan’s VAR limits blown twice during haywire Q1
Breaches add to the two regulatory backtesting exceptions sustained the previous quarter
DeepSeek success spurs banks to consider do-it-yourself AI
Chinese LLM resets price tag for in-house systems – and could also nudge banks towards open-source models
Industry fears Emir 3.0 fast model approval will cause delays
More model changes could be caught by proposed criteria for defining significance
How Citi moved GenAI from firm-wide ban to internal roll-out
Bank adopted three specific inward-facing use cases with a unified framework behind them
An AI-first approach to model risk management
Firms must define their AI risk appetite before trying to manage or model it, says Christophe Rougeaux
The impact of deterioration in rating-model discriminatory power on expected losses
The authors propose a means to estimate the effects on a portfolio’s expected credit loss created by underwriting model risks.
US banks seek to open vendors’ black box on green data
Inaugural Fed climate scenario analysis flags lack of transparency around third-party models
A study of China’s financial market risks in the context of Covid-19, based on a rolling generalized autoregressive score model using the asymmetric Laplace distribution
The authors construct a risk measurement model for the financial market during the Covid-19 pandemic, using data from the Shanghai Stock Exchange for empirical analysis.
How Ally found the key to GenAI at the bottom of a teacup
Risk-and-tech chemistry – plus Microsoft’s flexibility – has seen US lender leap from experiments to execution