Applied notes on risk, AI and modelling
An expanding research library on institutional risk modelling — covering model validation, IFRS 9 and IFRS 17, Solvency II and ORSA, liquidity and IRRBB, credit stress testing for ICAAP, climate and catastrophe tail risk, geopolitical risk, AI/ML model governance and AI-native risk platforms. Each article is written for risk professionals moving from theory to implementation.
Browse by topic
Climate Risk Governance for Banks: Board Oversight, Risk Appetite, Data and Controls
A bank can have sophisticated climate models and still manage climate risk poorly if ownership is unclear. Governance determines who approves the methodology, who owns the exposure, who.
How to Validate a Climate Risk Model: Calibration, Uncertainty, Scenarios and Governance
Climate risk models often combine several models rather than one. An emissions scenario may feed a climate model, which feeds a regional hazard model, which feeds an exposure and.
Climate Change and Catastrophe Models: The Challenge of Non-Stationarity
Catastrophe models generally combine three core components: hazard, exposure and vulnerability. Hazard describes the frequency and intensity of events. Exposure describes the assets, people.
RIME-X and Probabilistic Climate Risk Modelling: What Financial Risk Professionals Should Know
Financial institutions need climate information at a scale that can be linked to exposures, but comprehensive Earth system and impact models are expensive and cover only a limited set of.
Beyond Logic Trees: Bayesian Methods for Climate and Catastrophe Risk
Logic trees have been used for decades in probabilistic risk assessment because they provide a disciplined way to represent alternative assumptions. They are effective when the problem can.
Pricing Climate Risk: Cash Flows, Discount Rates, WACC and Asset Valuation
Climate risk can affect the value of an asset through several channels at the same time. A physical event may reduce revenue, damage assets and increase insurance costs. A transition shock.
Climate Stress Testing for Banks: Turning NGFS Scenarios into Capital and Risk Decisions
Climate stress testing is often weakened by a basic conceptual error: treating a scenario as a prediction. A forecast asks what is most likely to happen. A stress scenario asks what could.
Physical Climate Risk and Bank Lending: What Flood Events Reveal About Credit Risk
One of the most important lessons from recent banking research is that physical climate risk is granular. A flood does not affect every borrower in a region equally. Two firms separated by.
Integrating Climate Risk into Credit Risk Models: From Physical Hazards to PD, LGD and Lending Decisions
Adding a climate score to a borrower rating does not automatically create a climate-sensitive credit model. A credible approach requires an economic transmission mechanism. The question is.
Climate Risk Is Financial Risk: How Banks Can Move from Disclosure to Decision-Making
Climate risk becomes financially relevant when a change in weather, policy, technology or market behaviour alters cash flow, asset values, funding needs, collateral quality, insurance.
Actuarial Reserving and IFRS 17: Making Uncertainty Explicit
A practical treatment of actuarial reserving under IFRS 17 — triangles, chain ladder and Bornhuetter–Ferguson, process/parameter/model uncertainty and the risk adjustment.
AI and Machine-Learning Model Risk: A Validation Playbook
Independent validation of AI/ML models — leakage, drift, explainability, fairness, uncertainty, adversarial and out-of-distribution testing, and safe fallbacks.
Climate and Catastrophe Risk: Frequency, Severity and the Tail
A rigorous frequency-severity treatment of climate and catastrophe risk — Poisson and negative-binomial event counts, generalised Pareto tails, dependence, exposure and vulnerability.
Credit Risk Stress Testing for ICAAP: From Macro Scenarios to Capital Impact
How to build credit-risk stress tests for ICAAP — macro-satellite models, PD/LGD migration, concentration effects, and scenario design that survives supervisory review.
Geopolitical and War Risk: A Quantitative Modelling Framework
A defensible quantitative approach to geopolitical and war risk — state-transition models, Bayesian updating, event frequency-severity and honest treatment of scenario uncertainty.
IFRS 9 PD, LGD and EAD: Calibration and Validation in Practice
How to calibrate and validate IFRS 9 PD, LGD and EAD components — PIT vs TTC, downturn effects, cure and redefault, backtesting, staging and management overlays.
IRRBB and CSRBB: EVE, NII and Behavioural Models in Practice
A practical treatment of IRRBB and CSRBB — EVE and NII measurement, deposit-decay and prepayment behaviour, basis and optionality, and the validation view.
Liquidity Risk: LCR, NSFR and ILAAP Under Stress
A practical guide to LCR, NSFR and ILAAP liquidity modelling — behavioural cash flows, survival horizon, idiosyncratic and market-wide stress, and reverse stress.
The Model Validation Lifecycle: Independent Challenge from Concept to Retirement
A practical end-to-end walk-through of model validation — conceptual soundness, data, implementation, outcomes analysis, robustness, governance and remediation — aligned with SR 11-7 and EBA expectations.
Solvency II, ORSA and Economic Capital: Connecting Three Views Without Confusing Them
How to connect Solvency II SCR, ORSA and economic capital coherently — same risks, different lenses — without collapsing them into a single number that satisfies no one.
Beyond Historical VaR: Expected Shortfall, Stressed Calibration and Tail Dependence
Why historical VaR fails in crises, and how expected shortfall, stressed calibration and tail-dependence modelling produce capital numbers that hold up.
Frequency–Severity Modelling of Insured Catastrophe Losses
Why the compound Poisson frequency–severity decomposition still prices most catastrophe risk, and how to calibrate and stress it honestly for todays perils.
From Hazard to Loss: A Financial Framework for Physical Climate Risk
A calibrated hazard–exposure–vulnerability–loss chain that turns climate heat maps into balance-sheet numbers regulators and boards will accept.
The Future of Catastrophe-Risk Modelling: From Black Boxes to Transparent Frameworks
Why the next generation of catastrophe models will be judged on transparency and explainability, not just numbers — and how secondary perils are reshaping the modelling landscape.
NGFS Climate Scenario Translation: A Practical Guide to Climate Risk Modelling
A practical guide to translating NGFS climate scenarios into financial impacts using hazard-specific physical risk functions and explainable AI.
How AI-Native Risk Platforms Can Improve ICAAP, ORSA and Stress Testing
Explores how explainable AI, live calibration and integrated scenario engines close the gap between regulatory expectations and legacy static models.
From Fixed Assumptions to Calibrated Risk Engines
Why moving from hardcoded assumption sets to data-driven, versioned calibration pipelines strengthens governance and model performance.
Climate Physical Risk: Data Sources, Calibration and Validation by Hazard
A pragmatic walk-through of hazard-specific data pipelines for flood, wind and heat, including calibration choices and validation metrics.
IRRBB Modelling: Yield Curve PCA, Behavioural Assumptions and Delta NII
Building an IRRBB engine that combines PCA-decomposed yield curve shocks with defensible behavioural overlays for NII and EVE.
Cross-Asset Stress Testing for Banks and Insurers
Designing coherent macro-financial shocks that translate consistently into rates, credit, equity, property, FX and insurance liabilities.