top of page
empresario-analizando-los-datos-del-mercado-de-valores.jpg

Market Risk, FRTB & Artificial Intelligence

 

 

Programme Description

​

Market Risk is evolving rapidly. Traditional frameworks based on Value at Risk, static stress scenarios and periodic limit monitoring are increasingly being complemented by advanced analytics capable of detecting regime changes, identifying hidden risk concentrations, generating nonlinear stress scenarios and monitoring portfolios in near real time.

​

This programme provides a comprehensive and modern framework for Market Risk Management, FRTB, Stress Testing, Model Validation and Artificial Intelligence.

Participants will learn how classical market risk methodologies can be integrated with Machine Learning, Deep Learning, Generative AI, Reinforcement Learning, anomaly detection, graph analytics and autonomous AI agentsto build more dynamic, forward-looking and explainable risk management systems.

​

The programme covers the complete market risk lifecycle:

​

MARKET DATA → RISK FACTORS → VOLATILITY & REGIMES → PORTFOLIO EXPOSURES → FRTB / EXPECTED SHORTFALL → STRESS TESTING → MODEL VALIDATION → LIMITS → EARLY WARNING → AI MARKET RISK COPILOT

 

The objective is not to replace traditional quantitative finance. Rather, the programme shows how AI can enhance risk identification, scenario generation, portfolio revaluation, stress testing, model monitoring and decision support while maintaining robust governance, validation and human oversight.

​

Programme Objectives

 

By the end of the programme, participants will be able to:

  • Understand the architecture of a modern market risk management framework.

  • Apply the main components of FRTB, including the Standardised Approach and key elements of the Internal Models Approach.

  • Build and map market risk factors across complex trading portfolios.

  • Measure and compare Value at Risk and Expected Shortfall using both classical and AI-based approaches.

  • Model volatility and market regimes using econometric and Machine Learning techniques.

  • Analyse dynamic correlations, tail dependence and nonlinear risk interactions.

  • Apply Artificial Intelligence to discover hidden portfolio risk factors and concentrations.

  • Develop advanced risk models for interest-rate, FX, equity, credit spread, commodity and option portfolios.

  • Use AI surrogate models to accelerate large-scale portfolio revaluation.

  • Analyse liquidity-adjusted market risk, bid-ask widening, fire-sale effects and market depth.

  • Build geopolitical and cross-asset market stress scenarios.

  • Generate and rank large numbers of stress scenarios using AI.

  • Apply Reverse Stress Testing and Scenario Discovery to identify portfolio vulnerabilities.

  • Understand the market risk implications of algorithmic trading, crowded trades and AI-driven strategies.

  • Perform FRTB backtesting, P&L Attribution and advanced model diagnostics.

  • Validate Machine Learning and AI-based Market Risk models.

  • Develop dynamic limits, early-warning indicators and predictive limit-breach models.

  • Apply Generative AI to market intelligence, scenario analysis and risk reporting.

  • Design AI-assisted and multi-agent workflows for Market Risk Management.

  • Integrate model governance, validation, auditability and human oversight into AI-enabled risk processes.

 

Who Should Attend?

​

This programme is designed for professionals involved in Market Risk, Trading, Treasury, quantitative modelling and financial risk governance.

It is particularly relevant for:

​

  • Market Risk Managers and Analysts

  • Trading Risk and Treasury Risk Professionals

  • Traders and Structurers

  • Quantitative Analysts and Quant Developers

  • Model Validation and Model Risk Teams

  • Risk Analytics and Data Science Teams

  • Product Control and Independent Price Verification Teams

  • Internal Audit and Risk Control Functions

  • FRTB Implementation Teams

  • Regulators and Banking Supervisors

  • Financial Risk Consultants

  • Professionals involved in AI applications for financial markets

 

A working knowledge of financial markets, derivatives, statistics and risk management is recommended. Experience with Python, R or quantitative modelling is useful but not essential for understanding the overall framework.

 

 

 

fondo-azul-degradado-lujo-abstracto-azul-oscuro-liso-banner-estudio-vineta-negra.jpg

AGENDA
Market Risk, FRTB & AI

 

 

Module 1 — The AI-Powered Market Risk Architecture

​

From Trading Book to Real-Time Risk Intelligence

This module introduces the architecture of a modern Market Risk framework and explains how positions, market data, pricing models, risk factors, P&L, capital and AI analytics interact.

Key Topics

  • Trading Book architecture

  • Trading desks and portfolios

  • Market data

  • Risk-factor mapping

  • Pricing dependencies

  • Greeks

  • P&L

  • VaR and Expected Shortfall

  • Stress Testing

  • Capital

  • Limits

  • AI analytics

  • Real-time vs batch risk

RiskLab 1 — Build a Market Risk Digital Twin

Create a simulated trading book containing:

Rates + FX + Equity + Credit + Commodities + Options

and dynamically recalculate exposures, P&L and risk metrics under changing market conditions.

 

Module 2 — FRTB: Standardised Approach, IMA & Trading Book Capital

​

From Regulatory Framework to Practical Implementation

This module develops the core mechanics of FRTB and their implications for trading desks and capital.

Key Topics

  • Trading Book / Banking Book boundary

  • Trading Desk definition

  • Internal Risk Transfers

  • Sensitivities-Based Method

  • Delta

  • Vega

  • Curvature

  • Default Risk Charge

  • Residual Risk Add-On

  • Expected Shortfall

  • Liquidity Horizons

  • Modellable / Non-Modellable Risk Factors

  • Internal Models Approach

  • P&L Attribution

  • Backtesting

  • Standardised vs Internal Model approaches

RiskLab 2 — FRTB Capital Engine

Build a simplified:

SBM + DRC + RRAO

framework and compare it with an internal Expected Shortfall approach.

 

Module 3 — AI Market Data Intelligence & Risk-Factor Discovery

​

Detecting Risk Before Modelling It

Artificial Intelligence is applied to market data quality, anomaly detection and discovery of hidden risk factors.

Key Topics

  • Market data quality

  • Tick data

  • Missing observations

  • Stale quotes

  • Outlier detection

  • Alternative market data

  • Feature engineering

  • PCA

  • Dynamic PCA

  • Autoencoders

  • Isolation Forest

  • Representation Learning

  • Hidden factor discovery

RiskLab 3 — AI Risk Factor Discovery Engine

Starting with hundreds of financial time series, the AI engine identifies:

Relevant Factors → Clusters → Anomalies → Hidden Exposures

and compares the results with traditional PCA.

 

Module 4 — AI Volatility Intelligence & Market Regime Detection

​

Beyond Traditional GARCH

This module combines classical volatility models with modern AI techniques to detect changing market environments.

Key Topics

  • EWMA

  • GARCH

  • EGARCH

  • GJR-GARCH

  • Stochastic Volatility

  • Realised Volatility

  • Intraday Volatility

  • XGBoost

  • LSTM

  • Temporal Fusion Transformer

  • Hidden Markov Models

  • Markov Switching

  • Change-Point Detection

RiskLab 4 — AI Market Regime Detector

Automatically classify market conditions into:

đŸŸ¢ Low Volatility
đŸŸ¡ Transition
đŸŸ  High Volatility
đŸ”´ Crisis

and adjust risk parameters dynamically.

 

Module 5 — Dynamic Correlation, Dependence & Tail Risk with AI

​

When Correlations Break Down

This module develops modern approaches for modelling dependence structures during periods of market stress.

Key Topics

  • Dynamic correlations

  • DCC-GARCH

  • Gaussian Copula

  • Student-t Copula

  • Vine Copulas

  • Tail Dependence

  • Extreme Value Theory

  • Correlation Breakdown

  • Autoencoders

  • Nonlinear Dependence

  • Stress Correlation Matrices

RiskLab 5 — Correlation Breakdown Simulator

Compare:

Historical Correlation

vs

Dynamic Correlation

vs

Crisis Correlation

and measure the impact on a multi-asset portfolio.

​

Module 6 — Value at Risk & Expected Shortfall 2.0

​

From Historical Simulation to Predictive Distributions

This module integrates classical VaR methodologies with AI-based distributional forecasting.

Key Topics

  • Historical Simulation

  • Filtered Historical Simulation

  • Parametric VaR

  • Monte Carlo

  • EVT

  • Expected Shortfall

  • Bootstrap

  • Bayesian VaR

  • Quantile Regression

  • Quantile Random Forest

  • Gradient Boosting

  • Deep Quantile Networks

  • Distributional Forecasting

RiskLab 6 — Classical vs AI VaR & Expected Shortfall

Compare:

Historical Simulation

GARCH

EVT

XGBoost Quantile

Deep Quantile Model

under both normal and stressed periods.

 

Module 7 — Portfolio Mapping & AI Exposure Intelligence

​

Understanding Where the Risk Really Is

This module focuses on mapping market risk exposures across complex portfolios.

Key Topics

  • Risk-factor mapping

  • Cash-flow mapping

  • Greeks mapping

  • Look-through analysis

  • Basis Risk

  • Proxy Risk

  • Portfolio decomposition

  • Factor sensitivities

  • AI clustering

  • Exposure concentration

  • Hidden common factors

RiskLab 7 — AI Portfolio Risk Map

Create a heat map linking:

Risk Factor → Position → Desk → Product → P&L Contribution → Capital Contribution

 

Module 8 — AI Interest Rate & Fixed-Income Market Risk

​

From Curve Risk to Intelligent Yield-Curve Scenarios

Key Topics

  • Yield Curve Risk

  • Key Rate Duration

  • DV01

  • Convexity

  • PCA

  • Curve shocks

  • Basis Risk

  • Credit Spread Risk

  • Stochastic Rates

  • Neural Yield-Curve Forecasting

  • AI Scenario Generation

  • Portfolio Revaluation

RiskLab 8 — AI Yield Curve Risk Engine

Generate yield-curve scenarios using AI and measure impacts on:

Bonds + IRS + FRA + Swaptions

 

Module 9 — AI FX, Equity & Commodity Risk

​

Integrated Multi-Asset Market Risk

This module combines major market-risk asset classes within a common framework.

Key Topics

  • FX Risk

  • FX Forwards and Options

  • Equity Risk

  • Beta Instability

  • Commodity Risk

  • Oil

  • Metals

  • Energy

  • Cross-Asset Dependence

  • Volatility Regimes

  • Machine Learning

  • Cross-Market Contagion

RiskLab 9 — AI Multi-Asset Risk Engine

Analyse:

EUR/USD + Equity Index + Oil + Gold

under dynamic volatility and correlation stress.

 

Module 10 — Options, Nonlinear Risk & Volatility Surface Intelligence

​

Managing Nonlinear Trading Book Exposures

Key Topics

  • Delta

  • Gamma

  • Vega

  • Cross Greeks

  • Volatility Smile

  • Volatility Surface

  • Local Volatility

  • Stochastic Volatility

  • Jump Risk

  • Gap Risk

  • Neural Volatility Surfaces

  • Scenario Revaluation

  • Nonlinear P&L

RiskLab 10 — AI Options Risk Surface

Build a three-dimensional:

Underlying × Volatility × P&L

risk surface and compare Delta-Gamma approximation with full revaluation.

 

Module 11 — AI Portfolio Revaluation & Surrogate Risk Models

​

From Hours of Computation to Seconds

Key Topics

  • Full Revaluation

  • Scenario Grids

  • Pricing Bottlenecks

  • Surrogate Models

  • Neural Networks

  • Gradient Boosting

  • GPU Acceleration

  • Approximation Error

  • Greeks Approximation

  • Large-Scale Portfolios

RiskLab 11 — 100,000-Trade AI Revaluation

Compare:

Full Revaluation

vs

Delta-Gamma

vs

AI Surrogate Revaluation

across thousands of market scenarios.

 

Module 12 — Default Risk Charge & Credit Spread Market Risk

​

Linking Credit Migration and Trading Book Capital

Key Topics

  • FRTB DRC

  • Jump-to-Default

  • Credit Spreads

  • Transition Matrices

  • PD

  • LGD

  • Credit Migration

  • Concentration

  • Correlation

  • Monte Carlo

  • AI Transition Modelling

RiskLab 12 — AI Default Risk Charge Engine

Compare:

Transition Matrix Approach

vs

Machine Learning Migration Model

for DRC estimation.

 

Module 13 — Market Liquidity Risk, Bid-Ask & Fire-Sale Effects

​

The Risk Traditional VaR Does Not Capture

Key Topics

  • Market Liquidity

  • Bid-Ask Widening

  • Market Depth

  • Liquidity-Adjusted VaR

  • Price Impact

  • Slippage

  • Crowded Positions

  • Fire Sales

  • Liquidity Horizons

  • Nonlinear Liquidation Costs

  • ML Liquidity Forecasting

RiskLab 13 — AI Fire-Sale & Liquidity Stress Simulator

Simulate:

Position Size → Market Impact → Price Decline → Loss → Additional Liquidation

​

Module 14 — Geopolitical Market Risk & Cross-Asset Stress Testing

​

From Geopolitical Events to Trading Book Losses

Key Topics

  • Geopolitical Shocks

  • Sanctions

  • Tariffs

  • Energy Risk

  • Commodity Shocks

  • FX

  • Sovereign Spreads

  • Equity

  • Safe-Haven Assets

  • Cross-Asset Transmission

  • Narrative-to-Market Mapping

  • AI News Analytics

RiskLab 14 — Geopolitical Trading Book Stress

Model:

Geopolitical Escalation

→ Oil ↑
→ Inflation ↑
→ Rates ↑
→ FX moves
→ Equity ↓
→ Credit Spreads ↑ 

→ Trading Book P&L

 

Module 15 — AI Scenario Generation & Market Stress Testing

​

From a Few Scenarios to Thousands of Intelligent Scenarios

Key Topics

  • Historical Scenarios

  • Hypothetical Scenarios

  • Factor Shocks

  • Monte Carlo

  • EVT

  • Copulas

  • Generative Models

  • Scenario Clustering

  • Conditional Scenarios

  • Scenario Plausibility

  • Tail Scenarios

  • Multi-Asset Revaluation

RiskLab 15 — AI Market Stress Scenario Generator

Generate thousands of scenarios and rank them by:

Severity × Plausibility × P&L Impact

 

Module 16 — AI Reverse Stress Testing & Vulnerability Discovery

 

What Scenario Breaks the Trading Book?

Key Topics

  • Reverse Stress Testing

  • Loss Thresholds

  • Capital Thresholds

  • Limit Breaches

  • Bayesian Optimisation

  • Genetic Algorithms

  • Surrogate Models

  • Scenario Discovery

  • SHAP

  • Vulnerability Mapping

RiskLab 16 — Find the Scenario That Breaks the Desk

Set:

Trading Loss > €100 million

and identify the combination of:

Rates + FX + Equity + Spreads + Volatility + Correlation

that produces the breach.

 

Module 17 — Algorithmic Trading, Crowded Trades & AI-Driven Market Risk

​

The Risk Created by AI Itself

Key Topics

  • Algorithmic Trading

  • AI Trading Strategies

  • Feedback Loops

  • Momentum Amplification

  • Crowded Trades

  • Herding

  • Model Convergence

  • Flash Liquidity

  • Procyclicality

  • Agent-Based Simulation

  • Market Microstructure

  • AI Systemic Interaction

RiskLab 17 — AI Trading Feedback Loop Simulator

Simulate multiple agents following similar signals:

Signal → Trading → Price Move → New Signal → Additional Trading

and observe endogenous amplification.

 

Module 18 — FRTB Backtesting, PLA & AI Model Diagnostics

From Traffic Lights to Intelligent Diagnostics

​

Key Topics

  • VaR Backtesting

  • Kupiec Test

  • Christoffersen Test

  • Basel Traffic Light

  • Expected Shortfall Validation

  • P&L Attribution

  • RTPL vs HPL

  • Distribution Diagnostics

  • Walk-Forward Validation

  • Regime-Sensitive Backtesting

  • AI Anomaly Detection

  • Failure Clustering

RiskLab 18 — AI Backtesting Diagnostic Engine

Instead of simply identifying a failure, the system determines:

When + Where + Why

the model failed.

 

Module 19 — AI Market Risk Model Validation & Model Risk

​

Validating Models That Learn

Key Topics

  • Model Inventory

  • Model Materiality

  • Pricing vs Risk Models

  • Data Risk

  • Specification Risk

  • Calibration Risk

  • Benchmarking

  • Model Uncertainty

  • Distribution Shift

  • Model Drift

  • AI Explainability

  • Out-of-Distribution Detection

  • Challenger Models

  • AI Governance

RiskLab 19 — Challenger Model Validation

Compare:

GARCH

vs

XGBoost

vs

LSTM

across different market regimes and identify when each model becomes unreliable.

 

Module 20 — Dynamic Limits, Early Warning & AI Market Risk Appetite

​

From Static Limits to Predictive Risk Control

Key Topics

  • Desk Limits

  • Trader Limits

  • VaR / ES Limits

  • Greeks Limits

  • Stop Loss

  • Concentration Limits

  • Stress Loss

  • KRI

  • Dynamic Thresholds

  • Anomaly Detection

  • Predictive Breaches

  • Limit Utilisation

  • Escalation

RiskLab 20 — AI Early Warning Trading Desk Dashboard

Forecast:

Probability of Limit Breach

for future trading sessions and explain the drivers using SHAP.

 

Module 21 — Generative AI for Market Risk Intelligence

​

From News to Risk Factors

Key Topics

  • Large Language Models

  • Retrieval-Augmented Generation

  • Financial News

  • Central-Bank Communication

  • Earnings

  • Geopolitical Intelligence

  • Market Narratives

  • Risk-Factor Extraction

  • Scenario Generation

  • Research Summarisation

  • Position-Aware Reporting

  • Hallucination Controls

  • Source Traceability

RiskLab 21 — Market Risk Intelligence Copilot

The user asks:

“What are the main risks to our trading book today?”

The system analyses:

Positions + Limits + Market Data + News + Scenarios

and produces quantified exposures and risk explanations.

​

Module 22 — Autonomous AI Agents & the Future Market Risk Desk

​

From Risk Calculation to Intelligent Risk Management

The final module integrates Market Risk analytics into a controlled multi-agent architecture.

Key Topics

  • Autonomous AI Agents

  • Agent Orchestration

  • Market Data Agent

  • Position Agent

  • FRTB Agent

  • VaR / ES Agent

  • Stress Testing Agent

  • Limit Agent

  • Validation Agent

  • Reporting Agent

  • Human-in-the-Loop

  • Permissions

  • Audit Trail

  • Kill Switches

  • Model Governance

RiskLab 22 — AI-Powered Market Risk Control Room

Build:

MARKET DATA AGENT

↓

TRADING BOOK AGENT

↓

FRTB AGENT

↓

VaR / ES AGENT

↓

STRESS TESTING AGENT

↓

LIMIT AGENT

↓

VALIDATION AGENT

↓

MARKET RISK COPILOT

The system automatically generates:

Risk Dashboard

Limit Breaches

Top P&L Drivers

Stress Losses

FRTB Metrics

Model Warnings

Suggested Risk Actions

with final human approval.

bottom of page