Date of Award
2017
Document Type
Thesis
Mentor
Miroslaw Mystkowski
Abstract
Excerpt from Introduction
Seldom reward is absent from risk, and stock markets are a prime example. Stock markets across the world are viewed as profitable and risky at the same time. Companies have made a business out of forecasting these markets. Quantitative analysis companies use mathematicians, financial analysts, and computer scientists to compete in the stock market. The old days of floor trading have progressed towards high-frequency trading with supercomputers housed within the exchange. For example, the New York Stock exchange has created regulations for these companies so that there’s competitive equality. The computer’s power, length of cable to the exchange, and more has been standardized so that no single company will have an advantage with the exception to algorithms. Computers are delegated the buying and selling of stocks in the New York Stock exchange. A computer receives information from the market, decides an action in microseconds, and that decision gets sent to the exchange in milliseconds. From the computer’s perspective, the difference between microseconds and millisecond is significant. The company’s trading algorithms are secretive and protected, but their performance depends on time series analysis and machine learning theory.
Citation Information
Linzie, Andrew, "Financial Analysis with Artificial Neural Networks Short-term Stock Market Forecasting" (2017). Undergraduate Honors Theses. 6.
https://digitalcommons.gardner-webb.edu/undergrad-honors/6