Portfolio Risk Models

A portfolio risk model you can defend: clean inputs, a covariance estimator chosen on evidence, factor and specific risk split, stress scenarios, and a standing check of predicted against realised risk.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Then open the Marketplace in the app, search for Portfolio Risk Models, and click Install. It goes into your own files, and works offline from then on.

Before you start: this plan is general information to help you organise decisions about money. It is not financial, investment, tax or legal advice, and it does not recommend any product, investment or deal. Rules differ between countries, so check decisions that matter with a qualified, licensed adviser in yours. Investing and trading can lose money, and raising or managing other people's money is regulated in most countries.

What it installs.

1

Pillar

1

Area

50

Projects

203

Starter tasks

🧮 Quantitative Finance & Trading

Portfolio Risk Models

Building or using risk models: covariance estimation, factor and specific risk, stress scenarios and comparing predicted with realised risk.

  1. Inventory of risk numbers and the decisions they drive
  2. Holdings, prices and returns feed for the risk run
  3. Build, buy or blend decision for the risk model
  4. Risk measures glossary and house conventions
  5. Sample covariance baseline and its failure points
  6. Estimation window and half-life choice
  7. Benchmark and active risk definitions
  8. First one-page portfolio risk report
  9. Risk model documentation file
  10. Predicted versus realised risk baseline
  11. Daily risk run sanity checks
  12. Weekly risk note to the portfolio manager
  13. Monthly bias test and forecast accuracy log
  14. Monthly covariance change log
  15. Quarterly stress scenario library refresh
  16. Risk limit monitoring and breach log
  17. Risk input exceptions queue
  18. Quarterly recalibration review
  19. Annual risk model review and sign-off
  20. Shrinkage estimators for covariance
  21. Factor and specific risk decomposition
  22. Marginal and component contributions to risk
  23. Value at risk and expected shortfall methods compared
  24. Random matrix theory and eigenvalue cleaning
  25. Correlations in sell-offs study
  26. Reading list for risk model practice
  27. Fat tails and the normality assumption
  28. Liquidity-adjusted risk study
  29. Specific risk for assets with short histories
  30. Missing data and new assets in the covariance
  31. Currency risk treatment decision
  32. Choosing risk horizons for each decision
  33. Head-to-head test of two covariance estimators
  34. Risk report redesign from reader feedback
  35. Look-through concentration review
  36. First stress results at the investment committee
  37. Risk model change go-live with a parallel run
  38. Post-mortem of a realised risk surprise
  39. Vendor risk model renewal review
  40. Risk answers for an allocator due diligence questionnaire
  41. Risk model for a personal multi-asset portfolio
  42. First 90 days as a new risk analyst
  43. Handing a strategy's risk profile to the risk team
  44. Lightweight risk process for a solo researcher
  45. Covariance estimator showcase for a career changer
  46. Consistent risk across multiple horizons
  47. Tail dependence and copula scenarios
  48. Risk for options and other nonlinear positions
  49. Reverse stress test of the portfolio
  50. Regime-conditional covariance research

Where it starts.

Each project arrives with its purpose, milestones and first tasks written out. Here is how the first three read.

Inventory of risk numbers and the decisions they drive

Risk models get built for one purpose and quietly reused for five others, so a covariance tuned for daily limit checks ends up sizing a three-month strategic allocation. Listing every risk number your team produces, who reads it and what decision it feeds shows where one model is being stretched past what it was designed for.

First tasks

  • Collect the last month of risk reports and screenshots in one folder
  • List each risk figure with the person who reads it
  • Ask three readers what decision each figure changes for them
  • Flag every figure used at a horizon the model was not built for

Holdings, prices and returns feed for the risk run

Most bad risk numbers trace back to inputs: a stale price, a position booked twice, a security mapped to the wrong identifier. A single, versioned feed of positions, prices and returns with documented sources is the ground every later risk project stands on.

First tasks

  • Export today's positions and note where each field comes from
  • Match every position to one identifier and one return series
  • Write a data dictionary covering each column in the feed
  • Store the feed with a date stamp so any past run can be rebuilt

Build, buy or blend decision for the risk model

Commercial factor risk models give coverage, history and an audit trail on day one, while an internal model fits your universe and horizon and costs research time to keep alive. Writing down the trade-offs before a renewal or a hire forces the choice onto cost, coverage and fit instead of habit.

First tasks

  • Write the five requirements the risk model must meet
  • Request methodology documents from two risk model vendors
  • Estimate yearly people time for running an internal model
  • Record the decision and what would make you revisit it

Make it yours.

Install Portfolio Risk Models, then tell Aurum the date you're aiming for. It checks whether the plan fits the weeks you actually have.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Blueprints are general information to help you organise your own work. They are not medical, financial, legal or other professional advice. See the Terms.