Note 4 built the machinery for generating alpha; this note covers the other half of the institutional job: measuring, decomposing and governing risk — and the industry settings where that job is done. It runs from the most applied end of the course (the Australian superannuation system, retirement products and their embedded option guarantees) through the core quantitative risk technology (linear factor risk models and the Euler decomposition behind risk budgeting) to the governance layer that regulators now demand (model risk management), and finishes with the production frontier: contextual/nonlinear alpha models, factor timing, GPUs, XVA and machine learning. Less unified than the earlier notes by design — these are the four lecture-driven topics that round out a working quant's view of the industry.
By the end of this note you should be able to:
- describe the Australian superannuation system (structure, roles, fees, tax, YFYS test) and explain why retirement changes the definition of risk (sequence and longevity risk);
- price and hedge variable-annuity guarantees (GMAB/GMWB) by simulation plus root-finding;
- derive the factor-model covariance and compare the three calibration philosophies;
- compute MCR, CR and PCR tables, justify the Euler decomposition, and extend it to VaR contributions;
- structure a complete model-risk-management answer: definition, materiality, lifecycle, validation, roles;
- explain contextual modelling, nonlinear factor effects and factor timing, with their data-mining dangers;
- sketch the post-2008 XVA framework, the GPU computing model, and the ML/alt-data stack.
In this lesson
- Contents
- The big picture
The full lesson is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Portfolio Theory and Quantitative Investment syllabus.