Multi-agent LLM simulations testing algorithmic collusion and coordination breakdown in oligopoly markets.

Python replication of Calvano et al. (2020)

Q-learning oligopoly simulation (Calvano et al., 2020) — high-fidelity Python reproduction with ≥ 95 % accuracy

Evidence Bundle profile for RetailTech FTC algorithmic-pricing readiness. 12 sections aligned with FTC Section 5 + 2023 algo-pricing guidance + Operation AI Comply + EU Omnibus 2019/2161 + state retail AI laws.

This repository features a precise used car valuation algorithm that sequentially calculates depreciation and appreciation based on age, mileage, owner history, collisions, and brand reliability. The project focuses on applying specific business logic rules, such as mileage limits and ownership impact, to determine a final vehicle value.

Bias Coverage Lab for RetailTech personalized-pricing / recommendation / loyalty disparate-impact testing. 6 cohort dimensions, 4-band verdict, aligned with FTC Section 5 + NY ATBP + CA CPPA ADMT + EU AI Act Art 10/15.

AI Incident Card profile for RetailTech: pricing discrimination complaints, recommendation bias surfaces, biometric misidentification, refund discrim, dark patterns enforcement events. 12 event types + 8 reason codes + 4-band severity verdict.

Replication of Calvano et al. (2020) 'Artificial Intelligence, Algorithmic Pricing, and Collusion' (AER)

A real-time pricing engine that transforms a bar menu into a simulated stock market.

A portfolio of quantitative finance models developed during the JPMorgan Chase Job Simulation, featuring continuous-time commodity forecasting, algorithmic storage contract pricing, and dynamic programming for FICO score quantization.