Machine Learning Algorithms in Financial Markets; Evaluating the Impact of Machine Learning Algorithms on Financial Market Equality
Grant, Benjamin, School of Engineering and Applied Science, University of Virginia
JACQUES, RICHARD, EN-Engineering and Society, University of Virginia
Bloomfield, Aaron, University of Virginia
My STS research and technical research are related as both topics look to achieve the goal of making it easier to invest in the stock market and eliminating the current barriers of entry. My STS research focuses on the current state of the financial markets and how machine learning algorithms make it difficult for new investors to enter the market. My technical research created a machine learning algorithm that can be used to forecast the movement of a stock in the upcoming week. The algorithm will help reduce the risk of beginner investors as they can see how the stocks in their portfolio might move throughout the week and incentivize more investors to join the market.
My STS research analyzed the impact of financial algorithms and the current barriers to entering the stock market for new investors through the actor-network theory. In this paper, I looked at some of the challenges preventing new investors from entering the stock market and the equality of the financial markets. I found that lack of education and resources were the primary reasons preventing someone from entering the stock market. Additionally, I went through a case study showing that despite attempts to democratize the stock market, retail investors still struggle to have equal access. I concluded by suggesting that personal finance and stock market education should be prioritized in high school as well as providing a machine learning tool for risk assessment for beginner investors. By providing retail investors with tools needed to enter the stock market today, the stock market would benefit as a whole.
My technical project aimed to provide a basic machine learning tool for stock price prediction for beginner investors. The machine learning algorithm provides a forecast for a specific stock, given the stock’s previous 50 days of trading. Prior to the creation of the algorithm, several various machine learning techniques were tested to find the most accurate model. The algorithm uses a single layer recurrent neural-net to predict how a stock will move throughout the week. The user would then be able to track whether the stock price will go up or down throughout the week to better manage their risk.
Both my STS and technical research aimed at making it easier for beginner and new investors to become more informed and confident when they invest in the stock market. Through my STS research, I found that while machine learning algorithms have been helpful tools in all industries, they can contain biases and negatively impact certain demographics. In the case of financial algorithms, their exclusivity makes it difficult for beginner investors to do well in the market. It is important when creating a machine learning algorithm, to ensure that it is free of bias and does not negatively impact anyone. The creation of my technical project helped me understand the power engineers hold while also gaining new perspectives on the ethical decisions they must make. Not only is it important to make ethical decisions when creating machine learning algorithms, but it is also important to analyze the effect the algorithms have after they are implemented as well. By working on these projects, I was able to come up with a solution to an ethical issue that is ongoing.
BS (Bachelor of Science)
Stock Market, Machine Learning Algorithm, Financial Market
School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Aaron Bloomfield
STS Advisor: Richard Jacques