Free module · Arbitrage & prediction markets
Prediction Markets: Contracts, Probability and Evidence
Read what settles, score what you forecast, and keep price separate from belief.
binary contracts · resolution · probabilities · scores · calibration
The building blocks
A prediction-market share is a contract about a resolution rule. Its price can be compared with a probability only after payoff, timing and costs are clear.
- Read and encode the resolution rule.
- Connect payout to a forecast and score.
- Check calibration, dependence and settlement uncertainty.
Lessons in this module
- A dollar claim is not a news headline
- From probability to a decision price
- Conditional probabilities and contract dependence
- Brier score: measure the whole probability
- Log loss and overconfident mistakes
- Calibration bins and their uncertainty
- Resolution delay and capital lock-up
- A causal forecast research ledger
Practice and apply
- A dollar claim is not a news headline — 100 shares at $.62, outcome YES=$1, total costs $1. Find net profit.
- From probability to a decision price — p=.66, ask=.62, costs=.01/share, 100 shares. Find expected USD profit.
- Conditional probabilities and contract dependence — P(A and B)=.30 and P(B)=.40. Find P(A|B).
- Brier score: measure the whole probability — Forecasts .8 and .3, outcomes 1 and 0. Find mean Brier score.
- Log loss and overconfident mistakes — p=.8 followed by YES. Using natural log, find log loss.
- Calibration bins and their uncertainty — 70 YES results among 100 independent events. Find plug-in SE of frequency.
- Resolution delay and capital lock-up — Certain $1 payout in 73 days, simple annual opportunity cost .10, 365-day convention. Find present value.
- A causal forecast research ledger — Model Brier .18, baseline Brier .24 on matching events. Find Brier skill.
Work through the practice exercises · Quant development tools