Interior_Banner_Events

Exploring the Performance of Different Machine Learning and Statistical Models for Time Series Forecasting of Stock Closing Prices in the Standard and Poor's (S&P) 500

Jonathan Geffrard, Christian Binoya, and Dr. Abishek Verma, Department of Computer Science, New Jersey City University, 2039 JFK Kennedy Boulevard, Jersey City, New Jersey 07305

The behavior of the stock market is difficult to predict on so many measures because of its volatility and nonlinearity. Hundreds and thousands of traders including buyers and sellers congregate in these markets in order to transact and conduct business. No longer are the days where traders had to physically arrive at venues and other facilities in order to transact with other traders as most of these transactions are conducted electronically on various stock exchanges, some of which are the New York Stock Exchange, NASDAQ, and the Hong Kong Stock Exchange. With the advent of information technology, the frequency at which these transactions are being conducted as well as the volume of these transactions in a real-time environment are causing traders, stockbrokers, investment companies, banks, and other financial firms to reevaluate their approach in both fundamental analysis and technical analysis of the behavior of the stock market. With respect to the technical analysis of the stock market, there has been increasing research and interest in implementing machine learning as an investment strategy in predictive modeling and quantitative analysis of the stock market. This paper seeks to gently introduce and explore the performance of different machine learning and statistical models for time series forecasting of stock closing prices in the Standard and Poor’s (S&P) 500 (hereinafter referred to as “S&P 500”) as well as synopsize our findings in tandem with the available literature in anticipation for any future plans and/or afterthoughts.




Additional Abstract Information

Presenters: Jonathan Geffrard, Christian Binoya

Institution: New Jersey City University

Type: Poster

Subject: Computer Science

Status: Approved


Time and Location

Session: Poster 5
Date/Time: Tue 12:30pm-1:30pm
Session Number: 4031