Python Portfolio Analytics

The app, despite its early stage, is working properly. You may encounter, however, issues where datasets are incomplete or wrong (from both yfinance and alpha vantage).

Test locally in conda environment:

$ git clone https://github.com/simprecicchiani/iPortfolio.git
$ cd iPortfolio
$ conda env create -f environment.yml
$ conda activate iportfolio

Alternatively with pip in a virtual environment (python 3.9):

$ git clone https://github.com/simprecicchiani/iPortfolio.git
$ cd iPortfolio
$ pip install -r requirements.txt

Run the web app locally with

$ streamlit run app/main.py

Open either /tests/yf_test.ipynb (uses Yahoo Finance, slower, support multicurrency) or /tests/av_test.ipynb (requires a free Alpha Vantage API key).

The idea

Building a proper portfolio tracker. Featuring historical allocation, cash flows, real returns and more.

UX

[In] Transactions

Date	Ticker	Order	Price	Quantity	Fee
2019-10-01	CASH.USD	deposit	1	100000	0
2019-10-12	AAPL	purchase	234.52	88	35
2019-11-24	MSFT	purchase	148.3	250	25
2019-12-04	AAPL	sale	267.76	50	20
2020-01-06	FB	purchase	208	100	10
2020-01-25	CASH.USD	withdrawal	1	30000	0

[Out] Portfolio stats

UI

Streamlit web app

Built with

Python 3.9 and additional packages (leveraging pandas DatetimeIndex feature)

To Do

• Migration to Alpha Vantage
• Multicurrency portfolio (done in yf)
• Data sanity check
• Portfolio builder UI
• Portfolio dashboard

Further development

• Portfolio Technicals
• Lots of charts
• Indicators (Sharpe Ratio, Beta, VaR, etc)
• Sector Exposure (stocks only)