Blackrock Algothon 2019

• Dema Ushchapovskyy, Alex Jarisch, Gasim Gasim, Arjun Perumalla

• October, 2019

24 hour challenge of utilising alternative datasets such as Facebook data and weather data to extract alpha and beta from the financial markets.

Under time pressure, extracted the necessary data from Quandl and Refintiv, formalised the problem into a partially observable Markov decision process. Built an equity trading environment, with interchangeable reward functions such as Sharpe ratio and simple return. The environment allowed maximisation of alpha for any equity instrument given the data for the equity instrument.

Final Place: 3rd.