Detect and quantify wash trading

Outline an analytical method for identifying and measuring wash trading in BTC–USD and low-liquidity altcoin markets on a centralized exchange. Specify: a) signals derived from order and trade logs (self-matches, fast round-trips, mirrored size/price behavior across linked accounts, order-book position churn); b) graph-based heuristics for inferring shared control (common devices/IPs, on-chain funding connections) using privacy-preserving hashing and strict access controls; c) cutoffs that distinguish legitimate market making from manipulation, including precision/recall trade-offs; d) backtesting with synthetic injected wash trades and any available enforcement ground truth; e) a daily risk score with confidence intervals and calibration checks; and f) how to avoid penalizing genuine liquidity providers and how you would escalate cases to Compliance for review. Overview: The question assesses a candidate's abilities in fraud detection analytics, building features from order and trade logs, graph-based identity inference, model calibration, and backtesting for detecting market manipulation.