MarkowitzPortfolioOptimization

Computing a solution for the optimal mean-variance tradeoff (maximising Sharpe Ratio) of a portfolio according to MPT.

This repository contains code that allows you to extract the composition and performance of any exchange-traded fund and attempt to allocate its constituent assets differently, according to the mean-variance optimization framework.

Algorithms supported:

• Unconstrained optimization
• Constrained (sum weights = 1) optimization
• Short-selling constrained optimization

Blog-like Description:

https://towardsdatascience.com/beating-the-etf-portfolio-optimisation-using-python-and-some-linear-algebra-e48d0e0e44f

Instructions

Data/ETF/ should have two .csv files for the benchmark fund: one with performance and one with composition
Data/Stocks/ should have all the stock files downloaded from Kaggle

Required libraries:

• pandas
• numpy
• matplotlib
• time
• os
• shutil
• datetime
• (pdb)
• scipy