Statistical Arbitrage Research

A point-in-time universe, honest spread tests, half-life and hedge ratio studies, borrow, cost and capacity checks, crowding monitors and written stop rules for researching pairs, baskets and residual reversion.

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

Then open the Marketplace in the app, search for Statistical Arbitrage Research, 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

200

Starter tasks

🧮 Quantitative Finance & Trading

Statistical Arbitrage Research

Researching statistical arbitrage: pairs and baskets, mean reversion tests, capacity, crowding, costs and why most spreads stop working.

  1. Stat arb research charter and scope
  2. Written risk limits and stop rule for spreads
  3. Point-in-time universe for pairs and baskets
  4. Price adjustments and borrow data for spread work
  5. Spread research log and idea register
  6. Distance method pairs baseline replication
  7. Synthetic mean-reverting spreads to test the pipeline
  8. Cost, slippage and borrow assumptions sheet
  9. Sealed holdout for spread research
  10. Core reading on statistical arbitrage
  11. Cointegration tests on candidate pairs
  12. Half-life of mean reversion estimates
  13. Hedge ratio estimation methods compared
  14. Kalman filter dynamic hedge ratios
  15. Entry and exit thresholds on spread z-scores
  16. PCA residual reversion study
  17. Basket and ETF versus constituents spreads
  18. Short-term reversal in the cross section
  19. Copula and nonlinear dependence for pairs
  20. Weekly candidate spread review
  21. Monthly spread decay check
  22. Borrow availability and fee check
  23. Corporate event watch on active pairs
  24. Shadow portfolio of candidate spreads
  25. Crowding monitor for spread strategies
  26. Quarterly capacity estimate refresh
  27. Quarterly regime review of mean reversion
  28. Choosing a pairs selection method
  29. Sector-neutral versus market-neutral construction
  30. Price stop versus time stop on diverging spreads
  31. Position sizing by spread volatility
  32. Pairs book or residual portfolio decision
  33. Turnover reduction in a reversion signal
  34. Liquidity filter tightening test
  35. Promotion review for a spread strategy
  36. Replaying the August 2007 quant unwind
  37. Opening the holdout for a final spread test
  38. Stat arb research seminar presentation
  39. Year-end stat arb research retrospective
  40. Stat arb on a solo researcher's budget
  41. Academic pairs papers versus desk standards
  42. Crypto spread research and venue risk
  43. Futures calendar and inter-commodity spreads
  44. First 90 days on a stat arb team
  45. Inheriting another researcher's spread book
  46. Intraday stat arb with high-frequency data
  47. Capacity and impact model for a spread book
  48. Measuring crowding with factor and holdings overlap
  49. Post-mortem library of spreads that stopped working
  50. Event-conditioned mean reversion filters

Where it starts.

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

Stat arb research charter and scope

Statistical arbitrage covers everything from two-stock pairs held for weeks to residual baskets turned over intraday, and research drifts when nobody has written down which corner you are in. A one-page charter fixes the asset class, universe, holding period, capital assumption and the question you are trying to answer, so every later test is judged against the same frame.

First tasks

  • Write one sentence describing the spreads you intend to study
  • Pick the universe, holding period and instruments and write them down
  • List the approaches that are out of scope for this research
  • File the dated charter at the front of the research log

Written risk limits and stop rule for spreads

Spreads that revert most of the time can diverge violently the rest, and the losses arrive in clusters when many funds hold the same trade. Writing gross and net exposure caps, per-spread loss limits, sizing rules and a strategy-level stop rule before any paper or live run means the decision to cut is made calmly, not in the middle of a drawdown.

First tasks

  • Write the maximum gross exposure for one spread and for the whole book
  • Set a per-spread loss limit and a time stop in trading days
  • Write the drawdown that pauses the whole strategy
  • Review the limits before any new spread goes to paper trading @recurring(quarterly)

Point-in-time universe for pairs and baskets

Pairs drawn from today's index members quietly exclude every company that was delisted, merged or went bankrupt, and those are exactly the pairs that broke. A point-in-time universe with historical membership, delistings and liquidity at each date removes the largest single source of flattering stat arb backtests.

First tasks

  • Check whether your data source carries historical index membership
  • Load delisted securities and their final returns for the test period
  • Compute a trailing traded value filter as of each formation date
  • Plot the eligible count per date and explain any jumps

Make it yours.

Install Statistical Arbitrage Research, 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.