DSLab WNE UW

Aim of the project

The main goal of this project is to test a number of algorithmic investment strategies for high frequency data using machine learning tools. The analyzes will be carried out using data for the most common currency pairs and cryptocurrencies starting from 2009.

Algorithmic trading means an automatic investment strategy that generates a buy or sell signal based on real-time asset prices and various indicators calculated on their basis. Statistical tools can be used to identify the investment in what assets (pairs or groups of assets in general) is potentially profitable at a given moment or to determine exactly when and on what scale the purchase or sale transaction should be made. Algorithms can also determine the optimal way of performing a transaction at a given moment.

Algorithmic trading means an automatic investment strategy that generates a buy or sell signal based on real-time asset prices and various indicators calculated on their basis.

Before implementing an algorithmic investment strategy, it should be tested on historical data, validating its profitability, investment risk and profitability stability over time.

The analyzes will test various investment strategies that use statistical arbitrage, including pair trading, and arbitration related to adjusting asset prices after the announcement of macroeconomic data (event arbitrage), e.g. interest rate or inflation data in European Union and the United States. Various machine learning algorithms will be used as statistical tools, including text mining.