To test my knowledge of Python, and play around with my stock portfolio, I created from scratch a Voilà finance application for small investors in the stock market.
The application retrieves from a standard csv the list of stock owned, runs a simple analysis on the return based on the last 2 years performances for every stock and the overall weighted portfolio, comparison with various industry indexes, and then suggests an optimize new portfolio based on a Sharpe Ratio analysis. The suggestion can be personalized inputting a maximum admissible volatility, a maximum percentage for a single stock in the portfolio, and the timeframe to consider.
If a second file with a wish list of stock is uploaded, each of the stock is analysed on its own and its correlation and beta with the actual portfolio is calculated. A final Sharpe Ratio analysis suggest if and how much of these stocks should be added to the portfolio to maximize the ratio.
The application is targeted to non-coders and for this reason can be deployed as standalone through platforms as Heroku.
The entire code is accessible and downloadable through GNU General Public License on my GitHub profile in both the interactive and non-interactive version.
User Manual for Jupyter notebook version:
It requires, to fully work, 2 datasets (the paths need to be inserted in the first chunk of code, for ease it’s set up as my downloads folder):
- portfolio_path: csv with the actual portfolio, formatted as a standard download of the cvs from yahoo finance
- new_stock_path: csv with the new stocks to analyze in relation with the actual portfolio; it only requires the “Symbol” column
- portfolio_path: csv with the actual portfolio, formatted as a standard download of the cvs from yahoo finance
- new_stock_path: csv with the new stocks to analyze in relation with the actual portfolio; it only requires the “Symbol” column
and it does the following:
1) ANALYSIS PER STOCK: for every stock in the portfolio last price, gain according to the weighted purchase price across all the purchases, last 2 years price trend, last 20 days price trend with moving averages, beta and alpha according to single factor using SPY, volumes, last 2 years price trend, mean daily return (2 years and last 20 days), annualized return (2 years and last 20 days), distribution of returns
2) ANALYSIS PORTFOLIO: same as analysis per stock but for the overall portfolio
3) CORRELATION BETWEEN STOCKS: a simple correlation matrix between all the stocks in the portfolio
4) COMPARISON PER INDUSTRY: using various industry indexes ("XLC", "XLY", "XLP", "XLE", "XLF","XLV", "XLI", "XLB", "XLRE", "XLK", "XLU", "SPDV", "DMDV", "EEMD") calculates correlation and beta for last 400 days and last 100 days, shows graphically portfolio vs index, summary table at the end
5) SHARPE RATIO OPTIMIZATION: with 2 years data performs a sharpe optimization; it uses the logarithmic returns and a RFR = 0 (graphically I think that the max sharpe line becomes a curve in the log space, that’s why the tangency portfolio seems to be in a weird place, am I correct? It still looks weird and I can’t understand why)
6) SHARPE RATIO OPTIMIZATION (custom) [1],[2]: through widget allows the user to select the preferred parameters for the optimization
7) ANALYSIS OF NEW STOCK: similar to analysis per stock + correlation and beta with portfolio last 400 days and last 100 days, summary table at the end
8) SHARPE RATIO OPTIMIZATION with new stocks (custom) [1],[2]: using all the new stocks in the portfolio, performs a similar optimization as sharpe optimization (custom)
1) ANALYSIS PER STOCK: for every stock in the portfolio last price, gain according to the weighted purchase price across all the purchases, last 2 years price trend, last 20 days price trend with moving averages, beta and alpha according to single factor using SPY, volumes, last 2 years price trend, mean daily return (2 years and last 20 days), annualized return (2 years and last 20 days), distribution of returns
2) ANALYSIS PORTFOLIO: same as analysis per stock but for the overall portfolio
3) CORRELATION BETWEEN STOCKS: a simple correlation matrix between all the stocks in the portfolio
4) COMPARISON PER INDUSTRY: using various industry indexes ("XLC", "XLY", "XLP", "XLE", "XLF","XLV", "XLI", "XLB", "XLRE", "XLK", "XLU", "SPDV", "DMDV", "EEMD") calculates correlation and beta for last 400 days and last 100 days, shows graphically portfolio vs index, summary table at the end
5) SHARPE RATIO OPTIMIZATION: with 2 years data performs a sharpe optimization; it uses the logarithmic returns and a RFR = 0 (graphically I think that the max sharpe line becomes a curve in the log space, that’s why the tangency portfolio seems to be in a weird place, am I correct? It still looks weird and I can’t understand why)
6) SHARPE RATIO OPTIMIZATION (custom) [1],[2]: through widget allows the user to select the preferred parameters for the optimization
7) ANALYSIS OF NEW STOCK: similar to analysis per stock + correlation and beta with portfolio last 400 days and last 100 days, summary table at the end
8) SHARPE RATIO OPTIMIZATION with new stocks (custom) [1],[2]: using all the new stocks in the portfolio, performs a similar optimization as sharpe optimization (custom)