Margin & Prime Brokerage — how the margin call is computed

Every margin methodology — from the fixed percentages of Reg T to a prime broker's house model — is an answer to one question: if this client defaults now, and I liquidate over the next few days, how much do I lose?This track walks the whole landscape: rules-based margin, the scenario grids behind portfolio margin and SPAN, the add-ons where real houses differentiate, cross-margining and its correlation traps, what digital assets break, and how the engine is actually built and backtested. There's an interactive lab in every module, and the track connects to the Fixed Income material on VaR, SIMM and repo.

1. ⏱️

   Module 0Available

   Initial margin vs variation margin

   VM settles the past, IM pre-funds the close-out window — the default timeline, the margin period of risk, segregation, SIMM and MVA.

   🎛 Default-timeline simulator

2. 🧭

   Module 1Available

   The one question every margin model answers

   If this client defaults now, how much do I lose before I'm out? Risk measure, liquidation horizon, add-ons — and why Reg T over-margins hedged books.

   🎛 Reg T vs risk-based lab

3. 🎯

   Module 2Available

   Portfolio margin & the scenario grid

   The OCC/TIMS approach: shock spot and vol across a grid, fully reprice everything, charge the worst cell — the ancestor of every risk-based system.

   🎛 Interactive scenario grid

4. 📊

   Module 3Available

   SPAN & SPAN 2 — the futures standard

   The 16-scenario risk array, the charges that scanning misses (spreads, delivery, the short option minimum), and why SPAN 2 moves to VaR.

   🎛 Risk array explorer

5. 🧱

   Module 4Available

   House margin — where the add-ons live

   The base number is commoditized; concentration (√t), liquidity, gap, wrong-way and stress add-ons are where prime brokers differentiate.

   🎛 Add-on stack builder

6. 🔗

   Module 5Available

   Cross-margining & the correlation problem

   Offsets are only worth granting if the correlation survives the liquidation scenario — stressed correlations, offset caps, and cross-entity netting.

   🎛 Correlation stress lab

7. ₿

   Module 6Available

   Digital assets — what actually breaks

   24/7 markets, vol several times equity scale, venue and custody risk, stablecoin haircuts, and the weekend gap in the margin period of risk.

   🎛 Crypto scan-range calibrator

8. ⚙️

   Module 7Available

   Inside a margin engine

   The seven-stage pipeline from positions to margin call — reproducibility, caching, add-ons as config, and the millisecond pre-trade what-if API.

   🎛 Pipeline explorer

9. 🚦

   Module 8Available

   Backtesting the margin model

   Did losses ever exceed the margin held? Exception counting, Kupiec and Christoffersen tests, and the traffic-light zones — same machinery as VaR backtesting.

   🎛 Exception counter

10. 🔀

    Module 9Available

    Wrong-way risk

    Exposure that peaks exactly when the counterparty is most likely to default — specific vs general WWR, the monoline and crypto-collateral cases, and Basel's alpha.

    🎛 Wrong-way scenario lab

Educational analysis, not investment advice.