A 2021 "arbitrage" bot for PancakeSwap Prediction, revisited as an empirical study. 144k on-chain rounds: naive parimutuel edges of +15–25% are erased by informed late betting — a reverse favorite–longshot bias that faded as pools deepened. Dataset, bot, and analysis reproducible in 3 commands.
MGT 595 coursework
Using high frequency bitcoin quotes from the bitcoincharts website. 🍎
Why retail trading is a coin flip — an empirical study on 10 years of DAX 1-minute data.
Comprehensive Python analysis of Premier League betting market inefficiencies (2005–2024). Evaluates bookmaker biases, betting strategies, and market efficiency using statistical methods and Monte Carlo simulations.
Modelo predictivo en entornos de alta incertidumbre para el Mundial 2026: análisis estocástico que integra variables tácticas y financieras mediante simulaciones de Monte Carlo y modelos de Poisson.
Open research by Cold Code Labs: the skewness of football betting returns is a structural invariant inherited from the sport, not produced by pricing.
Sequel to football-prediction-ml: ML could not beat the closing line, so this scanner stops predicting and buys mispriced odds instead. +4% avg CLV over 224 paper bets.
Three from-scratch sports lines (World Cup, MLB, tennis) blind-tested as bettor AND book against real odds — an honest market-efficiency study on free public data. Start with python3 explore.py.
Python-based time-series analysis of stock market volatility and market efficiency in major emerging economies with GARCH-family models.
A pre-registered search for exploitable mispricing in prediction markets: what was ruled out, at what statistical power, and four apparent edges that turned out to be sampling artifacts. Write-up and analysis code.
Research-grade Polymarket BTC Up/Down 5m data pipeline, calibration analysis, hypothesis backtests, and execution robustness.
Semantically-grounded lead-lag networks in Polymarket prediction markets
Statistical testing of the Random Walk Hypothesis on Bitcoin (BTC-USD) using ADF, Autocorrelation, Runs Test, Variance Ratio (Lo-MacKinlay), and GARCH(1,1) modelling - Python, statsmodels, arch
Cycle 2 of a pre-registered BTCUSDT research line (1h substrate, 1h–2d holds). CLOSED: KILL-for-now — 0/240 configs pass all gates; lower costs let 40 clear G5 but it's drift not skill; held-out sealed; D9 shelved the line.
Event-study analysis of US government contract awards and stock-market reaction for defense and aerospace contractors. Tests semi-strong market efficiency using contract data and sector-matched (ITA) Benchmarking. Includes documentation and correction from SPY to ITA.
Market-efficiency study of Polymarket x Kalshi: 92,290 markets scanned, 215 apparent price gaps, 0 executable after audit. Ships as a visual Lab plus a live cross-platform pair terminal.
IC-Gated Deployment Framework (ICGDF): two-stage statistical gate for cross-sectional equity ML. Mean IC -0.0005, gate closed 0/12 folds, ECE<0.025 despite zero discrimination. Under review at Computational Economics (Springer).
Quantitative research on prediction-market efficiency: Liang-Kleeman Information Flow, cross-asset generalization tests, options-implied mispricing, and a live-deployed paper trader. Rigorous out-of-sample validation.
Applying quantum information-flow and rigorous quantitative methods to financial markets — an honest, out-of-sample investigation of market efficiency (options pricing, causality, factor analysis).
Add a description, image, and links to the market-efficiency topic page so that developers can more easily learn about it.
To associate your repository with the market-efficiency topic, visit your repo's landing page and select "manage topics."