MATLAB Financial Risk Simulation and Backtesting Training Course

Course overview

Market-risk teams are expected to explain not only a portfolio's current VaR, expected shortfall, sensitivities and stress loss, but also whether the models producing those figures can be trusted. Spreadsheet-based processes and disconnected scripts make it difficult to reproduce scenarios, refresh market data, test assumptions or demonstrate model performance to risk committees and auditors. This course gives finance professionals a disciplined MATLAB workflow for building, validating and communicating market-risk simulations and backtests.

Participants use MATLAB and its financial and statistical toolboxes to prepare return data, model volatility and dependence, generate Monte Carlo and historical scenarios, and calculate portfolio risk measures. They implement parametric, historical and simulation-based Value at Risk (VaR), expected shortfall and stress-testing methods; construct fixed-income, equity and options portfolios; and examine the effect of correlations, fat tails and changing volatility. The course also covers VaR backtesting with exception analysis, Kupiec proportion-of-failures testing, Christoffersen independence testing, traffic-light interpretation and clear model-performance reporting.

Teaching combines instructor-led model walkthroughs with guided MATLAB coding labs based on a multi-asset trading portfolio. Each participant builds a reusable risk-simulation and backtesting workflow, including documented MATLAB live scripts, scenario inputs, risk outputs, validation tests and management-ready charts. The final workshop requires participants to defend model choices, interpret backtest results and identify remediation actions for a simulated model-risk review.

The programme is designed for intermediate practitioners who already work with market data, portfolios or risk reports and need stronger MATLAB-based analytical capability. It is particularly valuable where teams need repeatable, reviewable risk calculations rather than one-off analysis.

Course objectives

By the end of this course, participants will be able to:

• Build MATLAB live scripts that import, clean and align multi-asset price and return series
• Calculate parametric, historical and Monte Carlo VaR and expected shortfall for a portfolio
• Estimate volatility and correlation inputs using EWMA, covariance matrices and GARCH models
• Generate correlated market scenarios using Cholesky decomposition and Monte Carlo simulation
• Revalue equity, bond and option positions across simulated and historical stress scenarios
• Design stress tests and reverse-stress scenarios with documented market-factor shocks
• Execute VaR backtests using exception counts, Kupiec POF and Christoffersen independence tests
• Produce a model-validation report with risk charts, test evidence, assumptions and remediation actions

Benefits of attending

Target competencies

Who should attend

• Market Risk Analysts — who need reproducible VaR, expected shortfall and stress-testing models
• Risk Managers — who oversee risk limits and must challenge model outputs and backtesting evidence
• Quantitative Analysts — who want to operationalise portfolio simulations and validation tests in MATLAB
• Treasury Risk Specialists — who analyse interest-rate, liquidity and trading-book exposures
• Financial Engineers — who model derivatives and need to incorporate revaluation risk into portfolio analysis
• Model Validation Analysts — who review market-risk methodologies, assumptions and performance tests

Requirements and prerequisites

Participants should be comfortable with MATLAB fundamentals, including variables, matrices, tables, scripts, functions, plotting and importing data from CSV or Excel files. They should understand basic portfolio concepts: returns, volatility, covariance, correlation, confidence levels, bonds, options and Value at Risk. Experience producing risk reports or analysing market data is helpful. Attendees need access to MATLAB with Financial Toolbox, Statistics and Machine Learning Toolbox, and Parallel Computing Toolbox where available. Prior experience with Monte Carlo coding, GARCH estimation, formal model validation or advanced derivatives pricing is not required; these are developed during the course.

Training methodology

The instructor demonstrates each method in MATLAB before participants implement it in guided coding labs. Short technical sessions explain the risk rationale behind covariance estimation, scenario construction, revaluation and backtesting; hands-on exercises then apply the methods to equity, bond and option positions. Participants work individually and in peer review groups to inspect assumptions, compare model outputs and challenge exceptions. The final day uses a model-risk case study in which participants complete a validation pack and define an implementation plan for their own reporting environment.

Course outline

Day 1: MATLAB Foundations for Market-Risk Data

• MATLAB Live Scripts, sections and reproducible risk-analysis workflows
• Importing market data from CSV and Excel using readtable and datastore
• Cleaning missing observations, corporate-action gaps and non-trading dates
• Converting price series into simple, log and excess returns
• Using timetables to align equities, rates, FX and volatility series
• Exploratory return diagnostics with histograms, qqplots and rolling statistics
• Portfolio data structures, position weights and market-value aggregation

Workshop: Participants construct a cleaned multi-asset timetable and a documented MATLAB live script that produces portfolio returns and initial diagnostic charts.

Day 2: Risk Measures and Volatility Modelling

• Value at Risk and expected shortfall definitions, horizons and confidence levels
• Historical simulation VaR using sorted portfolio profit-and-loss distributions
• Variance-covariance VaR with normal-distribution assumptions
• Exponentially weighted moving average volatility estimation
• Covariance and correlation matrix estimation for multi-asset portfolios
• GARCH volatility forecasting using econometric model functions
• Fat tails, skewness and distributional assumption testing

Workshop: Participants calculate and compare historical, parametric and EWMA-based VaR and expected shortfall for a trading portfolio.

Day 3: Monte Carlo Simulation and Portfolio Revaluation

• Random-number streams, seeds and reproducible simulation design
• Correlated return generation using Cholesky decomposition
• Multivariate normal and Student t scenario generation
• Full revaluation versus delta-normal approximation
• Equity, fixed-income and FX position revaluation under market scenarios
• Option valuation with Black-Scholes functions and implied-volatility shocks
• Simulation performance improvement with vectorisation and parallel computing

Workshop: Participants build a Monte Carlo engine that generates correlated scenarios, revalues a mixed portfolio and reports simulated VaR and expected shortfall.

Day 4: Stress Testing and VaR Backtesting

• Historical stress scenarios and market-event selection
• Hypothetical factor shocks for rates, spreads, equities, FX and volatility
• Reverse stress testing from loss thresholds to required market moves
• Rolling VaR forecasts and realised profit-and-loss alignment
• VaR exception analysis and Basel traffic-light zone interpretation
• Kupiec proportion-of-failures test implementation
• Christoffersen independence and conditional-coverage test implementation

Workshop: Participants run rolling VaR backtests, classify exceptions, execute Kupiec and Christoffersen tests, and produce a traffic-light summary.

Day 5: Model Validation Reporting and Workplace Application

• Risk-model assumptions, limitations and model-risk documentation
• Benchmarking historical, parametric and Monte Carlo model performance
• Sensitivity analysis for confidence levels, lookback windows and correlations
• Visualising losses, VaR forecasts, exceptions and stress results in MATLAB
• Creating management-ready tables and charts from MATLAB outputs
• Validation findings, issue classification and remediation actions
• Deploying a controlled risk-analysis workflow with script versioning and review checkpoints

Workshop: Participants complete a model-validation pack containing MATLAB code, risk results, backtest evidence, charts, findings and a workplace implementation plan.

Tools & standards covered

MATLAB, Financial Toolbox, Statistics and Machine Learning Toolbox, Parallel Computing Toolbox

A typical training day

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions