Artificial Intelligence in Finance

Chapter 6. AI-First Finance

A computation takes information and transforms it, implementing what mathematicians call a function….If you’re in possession of a function that inputs all the world’s financial data and outputs the best stocks to buy, you’ll soon be extremely rich.

Max Tegmark (2017)

This chapter sets out to combine data-driven finance with the machine learning approach from the previous chapter. It only represents the beginning of this endeavor in that, for the first time, neural networks are used to discover statistical inefficiencies. “Efficient Markets” discusses the efficient market hypothesis and uses OLS regression to illustrate it based on financial time series data. “Market Prediction Based on Returns Data” for the first time applies neural networks, alongside OLS regression, to predict the future direction of a financial instrument’s price (“market direction”). The analysis relies on returns data only. “Market Prediction with More Features” adds more features to the mix, such as typical financial indicators. In this context, first results indicate that statistical inefficiencies might indeed be present. This is confirmed in “Market Prediction Intraday”, which works with intraday data as compared to end-of-day data. Finally, “Conclusions” discusses the effectiveness of big data in combination with AI in certain domains and argues that AI-first, theory-free finance might represent a way out of the theory fallacies in traditional finance.

Efficient Markets ...