Portfolio Management System

This portoflio management system aims to manage risk by using following methodologies:

-Efficient Frontier using simulated portfolios

-Hierarchial Risk Parity

-Risk Neutral Portfolio against S&P500

Initial extraction of data, data cleaning and preperation

1.Select five random US stocks: {BYD, S, KO, ELF, COCO} and benchmark: S&P500

2.Download the data (Aug 21 to Aug 23) from YahooFinance (BYD.csv, KO.csv...)

3.Extract the closing price columns from each of the dataset

4.Merge the dataset into one

5.Calculate simple/log returns

6.Calculate the variance covariance matrix and continue with following methodologies

Findings

Efficient Frontier using Simulated Portfolios

Parameters for portfolio at maximum sharpe ratio:

	BYD	COCO	ELF	KO	S
Weight allocation:	10.24%	10.38%	57.83%	20.78%	0.78%

Number of Portfolios:	500
Risk Free Rate:	4.27%
Annualized Return:	85%
Annualized Volatility:	31%

Risk Neutral Portfolio against S&P500

Calculating risk neutral portfolio using cvxpy optimizer by creating a setup of minimization that the sum of the constituents (weighted) betas must add up to 0, in order to achieve market neutrality.

Formula to calculate the beta: $$\beta_{spxprice,i} = \sigma_{i, spxprice} / \sigma_{spxprice}^2$$

Risk Parity

Weight allocation:

BYD	COCO	ELF	KO	S
0.17191521	0.13620398	0.14523945	0.459502	0.08713936

Portfolio (ex-ante) annualized volatility is: 5.13%

Single constituent's volatility is: 1.71%

File reference

BYD, KO, S, ELF, COCO.csv - raw datasets

Efficient_Frontier_using_Simulated_Portfolios.ipynb - Efficient Frontier methodology

Risk Parity.ipynb - Risk Parity methodology

Risk Neutral Portfolio.ipynb - Risk Neutral methodology using cvxpy optimizer

Special thanks for @FilippoGuerrieri26 for providing cvxpy optimizer setup template