Free module · Arbitrage & prediction markets
Prediction Strategies: Logic, Sizing and Market Making
Move from a forecast to a defensible, fully accounted decision.
complete sets · logical bounds · sizing · quoting · portfolio risk
The building blocks
Prediction strategies require contract logic, feasible prices and exposure limits. A probability model supplies only one input.
- Prove the contract relation in every state.
- Subtract costs and account for uncertainty.
- Size and evaluate the entire execution process.
Lessons in this module
- Complete-set purchases and redemption
- Subset relations and executable bounds
- Bounds for joint and union events
- Binary Kelly sizing and estimation error
- Decision buffers for probability uncertainty
- Quoting revenue and adverse selection
- Event overlap and portfolio variance
- Evaluate the decision process, including failed fills
Practice and apply
- Complete-set purchases and redemption — 200 sets, asks .47 and .50, total costs $2, $1 payout per set. Find surplus.
- Subset relations and executable bounds — A implies B. Sell A at .61, buy B at .58, costs .01 per pair. Find minimum modeled surplus.
- Bounds for joint and union events — p_A=.70, p_B=.60. Find the lowest coherent joint probability.
- Binary Kelly sizing and estimation error — Known p=.60, price a=.50, no costs. Find full long-only Kelly fraction.
- Decision buffers for probability uncertainty — Lower probability scenario .59, ask .57, costs .01/share. Find conservative expected edge.
- Quoting revenue and adverse selection — 100 pairs, captured spread .04, adverse loss .025, fees .005 per pair. Find expected dollars.
- Event overlap and portfolio variance — Weights .5,.5; return SDs .2,.2; correlation .8. Find same-horizon portfolio variance.
- Evaluate the decision process, including failed fills — Completion probability .9, successful gain $5, failure loss $20, additional expected costs $1. Find attempt expectation.
Work through the practice exercises · Quant development tools