Skip to content
View AnshDani2004's full-sized avatar
1 follower · 1 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report AnshDani2004

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AnshDani2004/README.md

Ansh Hemang Dani

Market Microstructure · Volatility Modeling · Quantitative Research

Portfolio LinkedIn

About

Incoming MS Financial Mathematics student at the University of Chicago.
CS + Mathematics at Barrett, The Honors College, ASU.

I build market microstructure systems, volatility research tools, and Bayesian regime models, with an emphasis on reproducible claims and honest failure analysis.

Recently shipped two flagship projects:

  • a C++ limit order book and market-making simulator with Nasdaq ITCH replay, paired same-seed statistics, and queue-position diagnostics;
  • a Python Vol Surface Research Lab for SPY option-chain cleaning, SABR/Heston calibration, robustness testing, and model failure diagnostics.

Featured Work

⚙️ C++ Limit Order Book And Market Making Simulator
Price-time priority · QQQ ITCH replay · paired 30-seed statistics · queue diagnostics

📈 Vol Surface Research Lab
SPY IV surfaces · SABR/Heston calibration · robustness testing · failure diagnostics

🏆 IMC Prosperity 4
#194 algorithmic / #256 overall · top 1.4% of 18,800+ teams

📉 Bayesian Sequential Decision-Making Thesis
HMM filtering · CUSUM regime detection · adaptive risk control

Selected Results

  • Built a deterministic C++ limit order book with price-time priority, partial fills, cancel/replace logic, self-trade prevention, CSV replay, and direct external_execute support.
  • Processed 1,000,000 synthetic order events at about 3.7M events/sec on Apple M3.
  • Replayed 12,423 bounded QQQ Nasdaq ITCH messages through the matching engine.
  • Added naive versus Avellaneda-Stoikov market-making simulations with risk controls, terminal liquidation, paired 30-seed statistics, and queue-position diagnostics.
  • Built a reproducible Python volatility-surface research engine for SPY option-chain cleaning, forward extraction, OTM IV surface construction, static-arbitrage diagnostics, SABR/Heston calibration, and model failure analysis.
  • Reduced SABR median RMSE from 0.0190 to 0.0077 after deterministic expiry/liquidity filtering; showed global/per-expiry Heston underfit SABR despite synthetic Heston recovery and implementation sanity checks.
  • Placed #194 algorithmic / #256 overall in IMC Prosperity 4 out of 18,800+ teams.

C++ · Python · NumPy · SciPy · pandas · PyTorch · SQL · R · CMake · pytest · ruff · Docker

Pinned Loading

  1. cpp-limit-order-book-market-maker cpp-limit-order-book-market-maker Public

    Deterministic C++ limit order book and market-making simulator with ITCH replay, Avellaneda-Stoikov experiments, queue diagnostics, 10-seed fill-rate analysis, and artifact validation.

    Python

  2. vol-surface-research-lab vol-surface-research-lab Public

    Python volatility-surface research engine for SPY option-chain cleaning, SABR/Heston calibration, robustness testing, cross-snapshot validation, and failure-mode analysis.

    Python

  3. BSD_Thesis BSD_Thesis Public

    Honors Thesis work

    TeX

  4. prizecup-quant prizecup-quant Public

    Dixon-Coles + Elo forecasting pipeline for World Cup 2026 prediction markets. 217 predictions across 22 matches, Brier score 0.226.

    Python

  5. PRISM PRISM Public

    Production-grade framework for prediction market signal discovery using Bayesian inference, regime detection, and rigorous walk-forward backtesting. 122 unit tests, 85% code coverage.

    Python

  6. bayesian-market-filters bayesian-market-filters Public

    Bayesian state space filters (Kalman, HMM, Particle) for latent fair value and volatility regime estimation. Walk forward validated on live BTC/USDT hourly data, OOS Sharpe +0.71 after realistic 1.…

    Python