An easy-to-use web app to find crypto arbitrage opportunities and perform backtesting.
The goal for me was to learn about the streamlit library, a powerful tool to create interactive python apps. Also, it simplifies and automates many steps of my previous statistical arbitrage scripts (see this repo for more details how it works).
pip install -r requirements.txt (tested under python 3.9)
from the main folder start the app with streamlit run app.py
Alternatively, you can try out the app without any installation directly on the streamlit cloud. 🥳
-
Fetch all tradeable USDT-Coin pairs from the bybit futures market along with their historic price data.
-
Calculate cointegration statistics to find potential opportunities.
-
Plot price curves and statistics of a given pair. Also creates the csv file used for backtesting.
-
Perform backtesting (i.e. show the performance of the arbitrage strategy when applied on past data of that pair)
Expandable text boxes explain each step in more detail.
-
More automation. Loop over all potentially profitable pairs and find the best opportunities.
-
Create web app also for the execution part, i.e. the actual trading bot.