Factor Models

A point-in-time factor library you can defend: clear definitions, cross-sectional regressions, exposure and factor return reports, crowding and regime checks, and a record of every methodology change.

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

Then open the Marketplace in the app, search for Factor 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

196

Starter tasks

🧮 Quantitative Finance & Trading

Factor Models

Building and testing factor models: factor definitions, cross-sectional regressions, exposures, factor returns, crowding and how factors behave across regimes.

  1. Factor model purpose and scope statement
  2. Point-in-time security universe for factor tests
  3. Fundamental data lag and restatement rules
  4. Total return series with delisting returns
  5. Factor definition sheet for the first five factors
  6. Benchmarking against a public factor return library
  7. Standardisation and outlier treatment conventions
  8. Industry classification choice for neutralising factors
  9. Factor research workbook and naming conventions
  10. Monthly factor return update
  11. Monthly factor exposure report for a portfolio
  12. Quarterly factor definition review
  13. Factor crowding watch
  14. Fundamentals refresh and data break check
  15. Factor correlation and redundancy monitor
  16. Weekly factor performance note
  17. Annual factor model rebuild cycle
  18. Factor research reading queue
  19. Factor library change log and versioning
  20. Fama-MacBeth cross-sectional regressions by hand
  21. Portfolio sorts and long-short factor spreads
  22. Time-series regressions and pricing error tests
  23. Characteristics versus factor loadings study
  24. Mapping the factor zoo by family
  25. Information coefficient and signal decay profile
  26. Statistical factors from principal components
  27. Standard errors and significance bars for factor premia
  28. Choosing a value factor definition
  29. Momentum construction choices and crash exposure
  30. Quality factor composite design
  31. Low volatility and low beta factor construction
  32. Size factor and microcap exclusion decision
  33. Sector-neutral versus raw factor comparison
  34. Factor rebalance frequency and turnover study
  35. Mixing versus integrating multi-factor portfolios
  36. Factor model presentation to an investment committee
  37. Peer review of factor definitions and code
  38. Year-end factor performance retrospective
  39. Factor drawdown post-mortem
  40. Presenting factor research at a seminar or conference
  41. Factor tilts in your own fund holdings
  42. From fundamental analyst to factor researcher
  43. Audit of an inherited factor library
  44. Turning a PhD in empirical asset pricing into practitioner factor work
  45. Explaining factor exposures to a portfolio manager
  46. Factor behaviour across market regimes
  47. Factor timing study with a sceptical bar
  48. International and emerging market factor build
  49. Style factors across asset classes
  50. Time-varying factor exposures and conditional betas

Where it starts.

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

Factor model purpose and scope statement

A factor model built to explain risk looks different from one built to find return, and teams that never write down which they want end up arguing about definitions for months. A one-page scope statement fixes the question, the universe, the horizon and the users before any data is pulled.

First tasks

  • Write one sentence on what the factor model is for
  • List the universe, rebalance frequency and horizon you will use
  • Name the people who will read the output and their decisions
  • Record three uses the model is explicitly not built for

Point-in-time security universe for factor tests

Most impressive factor results come from a universe that quietly drops the companies that later failed. Building the universe as it stood on each historical date, with delisted names kept and listing rules applied as they were then, is the single biggest protection against survivorship bias in factor research.

First tasks

  • Pull historical index or exchange membership with start and end dates
  • Add delisted and acquired securities back into each month
  • Apply liquidity and minimum price filters as of each formation date
  • Chart the monthly name count and annotate every jump

Fundamental data lag and restatement rules

Book value and earnings are not known on the fiscal year end, they arrive weeks or months later and are sometimes restated. Fixing a reporting lag rule and using first-reported figures where you have them stops value and quality factors from using numbers nobody could have traded on.

First tasks

  • Find out whether your data source stores filing dates and first-reported values
  • Write the lag rule for annual and quarterly fields
  • Run a check that flags any value used before its filing date
  • Save the rule beside the factor definition sheet

Make it yours.

Install Factor 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.