In this project, we implemented the statistical arbitrage strategy with a machine learning algorithm called genetic algorithm, in brief we collected the historical data from Binance website of BTC and Ethereum for the past two years, the difference between each data point was a minute, which was 1 million data in total for each currency, the data was used for the genetic algorithm which trained on it, learnt it and created parameters that the statistical arbitrage require to do the mathematics to know when to buy a currency and sell the other simultaneously The user of the application will have to choose how much money he wants to use for the trades, and how much percent of it is he willing to invest in each trade after that the user can choose to start with the current parameters or optimize them for higher and better profit which might be a little slower though Our project is a web page built using Python, html, CSS, and flask web framework which provides useful tools and features that make creating web applications in Python easier. It also gives flexibility and it’s a very accessible framework for developers since to build a web application quickly using only a single Python file.
Statistical arbitrage is supposedly one of the most popular types of trading strategy. In this strategy, usually a pair of stocks are traded in a market-neutral strategy, it doesn’t matter whether the market is trending upwards or downwards, the two open positions for each stock hedge against each other.
Historically, Statistical arbitrage evolved out of the simpler pairs trade strategy, in which stocks are put into pairs by fundamental or market-based similarities. When one stock in a pair outperforms the other, the underperforming stock is bought long, and the outperforming stock is sold short with the expectation that under performing stock will climb towards its outperforming partner. Mathematically speaking, the strategy is to find a pair of stocks with high correlation, cointegration, or other common factor characteristics.
Simply put, given a normal distribution of raw data points z-score is calculated so that the new distribution is a normal distribution with mean 0 and standard deviation of 1. Having such a distribution ~ N(0, 1) is very useful for creating threshold levels. For example, in pairs trading, we have a distribution of spread between the prices of stocks A and B. We can convert these raw scores of spread into z-scores as explained below. This new distribution will have mean 0 and standard deviation of 1. To calculating z-score we use the formula: z = (x – mean) / standard deviation, where x is a raw data point and z is the z-score. Mean and standard deviation can be rolling statistics for a period of ‘t’ days or minutes or time intervals. Using these concepts of moving averages and z-score we create the entry points for Pairs Trading.
- Entry point
- Upper Threshold
- Lower Threshold
- Exit point
- STOP LOSS
- TAKE PROFIT
- TIME OUT
It works on Python 3.9.0
git clone https://github.com/AHMADSK1997/statistical-arbitrage.git
pip install -r requirements.txt
python app.py
Connect to http://localhost:5000
