Python Portfolio Analysis

Contents

• Overview
• Data Preparation
• Quantitative Analysis
  • Performance Analysis
  • Risk Analysis
• Sharpe Ratios
• Custom Portfolio
  • Data Preparation
  • Re-run the Analysis, Comparing My Portfolio

Overview

I was tasked with analysing a number of financial portfolios across the areas of:

• volatility
• returns
• risk
• Sharpe ratios

I created an analysis notebook that analyzes and visualizes the major metrics of the portfolios across all of these areas, and determines which portfolio outperformed the others.

My data set included a number of different portfolios.

The three main tasks were to:

1. Read in and wrangle returns data
2. Determine success of each portfolio
3. Choose and evaluate a custom portfolio

Data Preparation

For the data preparation I was required to do the following:

• Use Pandas to read each CSV file in as a DataFrame.
• Detect and remove all null values.
• Convert the S&P 500 closing prices to daily returns.
• Join the Whale Returns, Algorithmic Returns, and the S&P 500 Returns into a single DataFrame with columns for each portfolio's individual returns.

See the single DataFrame below:

Quantitative Analysis

In this section, I was required to calculate and visualize performance and risk metrics for the portfolios.

Performance Analysis

Daily Returns

Cumulative Returns

Risk Analysis

Box Plot

Standard Deviation

Portfolios Riskier than S&P TSX 60

Annualised Standard Deviation

Rolling Standard Deviation with a 21-day Window

Calculate Correlation

Plot Correlation

Beta

I was required to calculate and plot Beta for a chosen portfolio (algo 1) and the S&P 60 TSX.

Sharpe Ratios

Annualised Sharpe Ratios

Sharpe Ratios Bar Plot

Determine whether the algorithmic strategies outperform both the market (S&P TSX 60) and the whales portfolios

I found as follows:

Custom Portfolio

Data Preparation

I created a custom portfolio of Cardano, Bitcoin and Ethereum.

I used the Google Finance function to choose a portfolio, then downloaded the data needed as CSV files.

In my notebook I read in the CSVs and joined them as a single DataFrame, as below:

I calculated the weighted returns for the portfolio assuming an equal number of shares for each stock:

Then I added these to my DataFrame containing all portfolios:

Re-run the Analysis, Comparing My Portfolio

Annualised Standard Deviation

Rolling Standard Deviation with a 21-day Window

Calculate Correlation

Plot Correlation

Beta

I was required to calculate and plot 60-day Rolling Beta for My Portfolio compared to the S&P 60 TSX.

Annualised Sharpe Ratios

Sharpe Ratios Bar Plot

My Portfolio Compared to the Market

For my portfolio, I found as follows: