國立高雄大學圖資館 |

Language: English

Back

An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price.

Record Type:	Electronic resources : Monograph/item
Title/Author:	An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price.
Author:	Abdelkader, Fadi.
Published:	Ann Arbor : ProQuest Dissertations & Theses, 2023
Description:	147 p.
Notes:	Source: Dissertations Abstracts International, Volume: 85-02, Section: B.
Notes:	Advisor: Kabia, Milton.
Contained By:	Dissertations Abstracts International85-02B.
Subject:	Finance.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30418556
ISBN:	9798379956691

An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price.
Abdelkader, Fadi.

An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price. - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 147 p.

Source: Dissertations Abstracts International, Volume: 85-02, Section: B.

Thesis (Ph.D.)--Northcentral University, 2023.

This item must not be sold to any third party vendors.

Automated Trading Systems (ATSs) are increasingly used in the stock markets due to their speed and accuracy. However, their use poses a risk of contributing to market crashes, and there is no standard approach to evaluating their decision-making process or performance. The study proposed a standardized approach for evaluating ATSs by examining the relationship between the algorithms' predicted price and predicted risk and the constructs of profitability and validity. The study introduced the Multiple-Attribute Utility Theory-based framework for consistently assessing ATS performance and validity for market trading. This study used a quantitative experimental approach to propose a framework for evaluating stock prediction algorithms and their potential to mitigate flash market crashes. The study analyzed eight machine learning models in 42 experiments driven by a 20-year dataset of ten US stocks using the Microsoft Azure cloud AutoML service. Experimental reliability and validity were investigated using statistical techniques such as test-retest, split-half, and interrater reliability. The results indicated that ATS trading significantly improves profitability compared to a buy-and-hold strategy. Further, the results reveal a significant difference in accuracy and profitability among algorithm classes, with ensemble algorithms outperforming individual models but being more compute-intensive. Finally, the results show that ATSs cannot detect the signals leading to flash crashes, highlighting the need for additional market controls. The research provides insights into ATSs' fairness, the profitability-cost trade-off, their impact on liquidity and economic stability, and the vulnerability of financial markets without controls to prevent flash crashes. The results of this study contributed to a greater understanding of ATSs and their impact on stock markets, paving the way for improved and safer stock markets for traders and economies.

ISBN: 9798379956691Subjects--Topical Terms:

183252
Finance.
Subjects--Index Terms:

Automated Trading Systems

An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price.
LDR:03217nmm a2200397 4500 001 655751
005 20240414211921.5
006 m o d
007 cr#unu||||||||
008 240620s2023 ||||||||||||||||| ||eng d
020 $a 9798379956691
035 $a (MiAaPQ)AAI30418556
035 $a AAI30418556
040 $a MiAaPQ $c MiAaPQ
100 1 $a Abdelkader, Fadi. $3 966813
245 1 3 $a An Evaluation of the Accuracy and Profitability of Machine Learning Algorithms in Predicting Stock Price.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 147 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-02, Section: B.
500 $a Advisor: Kabia, Milton.
502 $a Thesis (Ph.D.)--Northcentral University, 2023.
506 $a This item must not be sold to any third party vendors.
520 $a Automated Trading Systems (ATSs) are increasingly used in the stock markets due to their speed and accuracy. However, their use poses a risk of contributing to market crashes, and there is no standard approach to evaluating their decision-making process or performance. The study proposed a standardized approach for evaluating ATSs by examining the relationship between the algorithms' predicted price and predicted risk and the constructs of profitability and validity. The study introduced the Multiple-Attribute Utility Theory-based framework for consistently assessing ATS performance and validity for market trading. This study used a quantitative experimental approach to propose a framework for evaluating stock prediction algorithms and their potential to mitigate flash market crashes. The study analyzed eight machine learning models in 42 experiments driven by a 20-year dataset of ten US stocks using the Microsoft Azure cloud AutoML service. Experimental reliability and validity were investigated using statistical techniques such as test-retest, split-half, and interrater reliability. The results indicated that ATS trading significantly improves profitability compared to a buy-and-hold strategy. Further, the results reveal a significant difference in accuracy and profitability among algorithm classes, with ensemble algorithms outperforming individual models but being more compute-intensive. Finally, the results show that ATSs cannot detect the signals leading to flash crashes, highlighting the need for additional market controls. The research provides insights into ATSs' fairness, the profitability-cost trade-off, their impact on liquidity and economic stability, and the vulnerability of financial markets without controls to prevent flash crashes. The results of this study contributed to a greater understanding of ATSs and their impact on stock markets, paving the way for improved and safer stock markets for traders and economies.
590 $a School code: 1443.
650 4 $a Finance. $3 183252
650 4 $a Computer science. $3 199325
653 $a Automated Trading Systems
653 $a Evaluation framework
653 $a Flash crash
653 $a Machine learning
653 $a Stock market
690 $a 0800
690 $a 0508
690 $a 0984
710 2 $a Northcentral University. $b School of Technology. $3 942603
773 0 $t Dissertations Abstracts International $g 85-02B.
790 $a 1443
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30418556