Informed Trading Support for the Amateur Investoron the New York Stock Exchange

摘要

While a large number of data-powered investment solutions exist, the majority are targeted at enterprise-scale wealth generation. Wherein high capital, real-time market access, instant execution, and high-performance infrastructure are all common prerequisites for implementation and success of most wealth management and trading strategies. Amateurs, however, are often relegated to more manual approaches, frequently including technical analysis (trendlines), study of chart patterns and educated guesses/risks. In this work, we aim to harness the high-volume and high-velocity attributes of Big Data source offered by the New York Stock Exchange by providing a novel system for investment decisions that can support the amateur investor. Rather than performing high-frequency trading approaches, we use easily accessible public data and machine learning to automate chart pattern analysis in a way that suggests weekly trading as a more practical approach in the context of nonprofessional investors. We show that the system outperforms uniform investment over the course of a year, demonstrating potential Big Data-powered wealth generation for individuals outside the financial sector.