A collection of quantitative finance methods and algorithms implemented in Python.
monte_carlo_simulations/
Geometric Brownian Motion (GBM) simulations comparing price paths with and without parameter uncertainty. Demonstrates how estimation error in drift parameter affects confidence intervals.
trade_calculation_stochastic_rounding/
Portfolio rebalancing with stochastic lot rounding. Rounds trade sizes to valid lot sizes while preserving expected values through probabilistic rounding.
regression_methods/
Portfolio weight optimization using different regression approaches:
- OLS (Ordinary Least Squares)
- Ridge (L2 regularization)
- Lasso (L1 regularization)
- Elastic Net (L1 + L2)
index_tracking/
Jupyter notebooks demonstrating index replication and sparse portfolio optimization.
pip install -e .- Python 3.9+
- numpy, pandas, scipy, matplotlib
MIT