Bitcoin Systematic Trading Algorithms in the Cloud: Challenges and Opportunities

Abstract

The growth of Big Data is concurrent with the expansion and democratization of cloud computing. The ability to access a super computer is now available to any user with a credit card. Wide adoption has fueled the rise of many new industries and disruptive technologies. Building competitive, highly performing systematic algorithms is within reach for even small, bootstrapped organizations, but the availability of cheap computing power alone is insufficient to be highly successful. What matters is the amount of data an organization can access and analyze in a time sensitive fashion. We analyze the intersection of Big Data analytics and cloud computing with a focus on creating investment algorithmic strategies to trade different cryptocurrencies. It is now possible to access high quality, inexpensive time series data sets. However, the exponential increase in the size of data sets, as well as the number of data sets required makes analysis infeasible for even very powerful desktop computers. Specifically, this paper will delve into the application of Machine Learning, its unique demands, and its application to modeling cryptocurrencies through the use of Big Data and cloud computing leveraging AWS and PaperSpace. We will explore the unique approach of using a circuit of ML algorithms to improve predictive power and its dependence on cloud computing structure.