Skip to content

A high-performance Monte Carlo pricer for Exotic Derivatives (Asian & Barrier Options) featuring a hybrid architecture (C++17, Python/Numba, and Pybind11).

License

Notifications You must be signed in to change notification settings

Rakeshks7/Exotic-Quant-Engine

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Exotic Quant Engine

Language License Status Domain

Overview

Exotic Quant Engine is a production-grade Monte Carlo simulation framework designed to price path-dependent "Exotic" options (Asian and Barrier options).

Unlike standard vanilla option pricers, this project demonstrates a Hybrid Architecture widely used in Top Tier Investment Banks:

  1. Prototyping: Python with Numba JIT acceleration for rapid development.
  2. Production: C++17 for low-latency execution and memory management.
  3. Binding: Pybind11 to expose the C++ engine to Python for analysis.

Key Features

  • Path-Dependent Pricing: Handles Arithmetic Asian Options (average price) and Barrier Options (Knock-Out/Knock-In).
  • Variance Reduction: Implements Antithetic Variates to reduce standard error and improve convergence speed by ~50%.
  • Performance Optimization: * Python loop bypassing via Numba (Just-In-Time compilation).
    • Manual memory management and vectorization in C++.
  • Sensitivity Analysis: Numerical calculation of Greeks (Delta, Gamma, Vega) via Finite Difference Methods.
  • Validation: Pricing benchmarked against the Kemna-Vorst analytical approximation for Geometric Asian options.

Technical Deep Dive

The Structuring Challenge Structuring involves creating bespoke financial products to meet specific client hedging needs.

Asian Options: Used by corporations to hedge currency/commodity exposure over a period (smoothing out volatility).

Barrier Options: Used to cheapen the cost of protection (the option disappears if the market moves too far in your favor).

Disclaimer

This project is for educational and portfolio purposes only. It is not intended for use in real money trading. The pricing models implemented here rely on assumptions (Geometric Brownian Motion) that may not fully reflect real-world market dynamics (jumps, stochastic volatility).

About

A high-performance Monte Carlo pricer for Exotic Derivatives (Asian & Barrier Options) featuring a hybrid architecture (C++17, Python/Numba, and Pybind11).

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published