Skip to content

liamos7/quant-theory

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

40 Commits
 
 
 
 
 
 

Repository files navigation

Quantitative Finance: Theory and Applications

A collection of quantitative finance projects exploring derivatives pricing, risk management, and trading strategies. Combines mathematical theory with practical implementations in Python.

Author: Liam O'Shaughnessy
Background: Princeton Physics (Senior) | HEP Research at CERN
Contact: email | LinkedIn


Theoretical Foundation

I've compiled comprehensive notes covering the mathematical foundations of quantitative finance:

  • Stochastic calculus and martingale theory
  • Derivatives pricing (Black-Scholes, exotic options)
  • Interest rate and credit models
  • Portfolio optimization and risk management
  • Numerical methods (Monte Carlo, finite difference)
  • Machine learning applications

These notes synthesize material from Shreve, Bjork, and academic papers.


Projects

Objective: Model implied volatility dynamics using local and stochastic volatility frameworks.

Key Techniques:

  • Implied volatility surface extraction from SPY options
  • SVI/SSVI parametrization with arbitrage detection
  • Dupire local volatility implementation
  • Heston stochastic volatility calibration (Carr-Madan FFT and Lewis methods)
  • Volatility risk premium estimation

Results: Identified and corrected arbitrage violations in raw SVI, calibrated Heston, estimated vol risk premium.

Tech Stack: Python, NumPy, SciPy, Plotly, yfinance

View Project →


Objective: Understand how historical volatility and implied volatility behave with stock behavior around earnings announcements.

Key Techniques:

  • Historical Volatility (HV) baseline vs. Event-driven Realized Volatility
  • "Earnings multiplier" analysis ($RV_{earnings} / \sigma_{baseline}$) across GICS sectors
  • VIX regime conditioning
  • Directional bias and "Win Rate" statistical testing
  • IV analysis (coming soon with better historical data)

Results: Identified that the Communication Services sector exhibit the highest shock multiplier (~4.2x), while Energy displays the lowest (~2.0x). Confirmed that earnings multipliers contract during High-VIX regimes due to volatility normalization.

Tech Stack: Python, Pandas, Matplotlib, Seaborn, yfinance

View Project →


Simulation of delta-gamma hedging with transaction costs and comparison to theoretical costs.


Tech Stack

  • Languages: Python, C++ (for performance-critical components)
  • Libraries: NumPy, SciPy, Pandas, Matplotlib/Plotly
  • Data: Yahoo Finance, CBOE, Quandl
  • Tools: Jupyter, Git, VS Code

Background

I'm a Princeton physics senior with research experience in high-energy experimental physics at CERN, focusing on machine learning applications for particle detection and classification. I became interested in quantitative finance while exploring applications of stochastic processes and statistical methods to financial markets.

These projects represent my self-directed learning in mathematical finance, combining rigorous theory with practical implementation.


Contact

Interested in collaborating or have feedback? Reach out:


References

Key resources used in developing these projects:

  • Shreve, S. (2004). Stochastic Calculus for Finance II
  • Bjork, T. (2009). Arbitrage Theory in Continuous Time
  • Koralov and Sinai (1992). Theory of Probability and Random Processes
  • Schilling and Partzsch (2010). Brownian Motion
  • Gatheral, J. (2006). The Volatility Surface
  • Taleb, N. (1997). Dynamic Hedging

About

A collection of quantitative finance projects exploring derivatives pricing, risk management, and trading strategies. Combines mathematical theory with practical implementations in Python.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors