Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification
Stock market manipulation detection is important for both investors and regulators. Being able to detect stock manipulation and preventing it gives investors the confidence in the market fairness and integrity. It also helps maintaining liquidity of the stocks and market efficiency. Implementing dat...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613419126915072 |
|---|---|
| access_status_str | Open Access |
| author | Youssef, Sarah |
| author_browse | Youssef, Sarah |
| author_facet | Youssef, Sarah |
| author_sort | Youssef, Sarah |
| collection | Thesis |
| description | Stock market manipulation detection is important for both investors and regulators. Being able to detect stock manipulation and preventing it gives investors the confidence in the market fairness and integrity. It also helps maintaining liquidity of the stocks and market efficiency. Implementing data mining algorithms in manipulation detection is a relatively recent technique but in the past few years there has been an increasing interest in it's applications in this domain. The benefit of monitoring manipulative trade behavior is that it can be implemented on live feed of stock data, which saves a lot of time in detecting stock price manipulation. This research implements machine learning algorithms in detecting trade manipulations where trade behaviors artificially impact the National Best Bid and Offer (NBBO) of traded stocks. Research methodology implemented is based on feature extraction using signal analysis, taking advantage of the similarity between physical signals measured by machines and raw financial data. Accordingly, Continuous Wavelet Transform (CWT) is applied on actual manipulation data for feature extraction, Principal Component Analysis (PCA) and factor analysis are used for dimensionality reduction and then Machine Learning Classifiers are trained and tested. Tick Bid/Ask Price and volume data of actual 15 manipulation cases published by the Security Exchange Center (SEC) was extracted from an online interface and labeled accordingly. This data was then used to train, and test 3 different classification models (XGBoost, KNN & SVM) and the outcome was compared accordingly. Results showed that introducing continuous wavelet transform enhances model accuracy, it increased precision results tremendously, while reducing recall values slightly. Adding PCA, reduced run time greatly, yet reduced the quality of some models prediction. Out of the three classifiers XGboost & KNN are showing the highest performance. |
| format | Thesis |
| id | oai:fount.aucegypt.edu:etds-2578 |
| institution | American University in Cairo (Egypt) |
| last_indexed | 2026-06-10T12:35:50.652Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from AUC Knowledge Fountain — bepress |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| publisher | AUC Knowledge Fountain |
| publisherStr | AUC Knowledge Fountain |
| record_format | dspace |
| source_str | AUC Knowledge Fountain — bepress |
| spelling | oai:fount.aucegypt.edu:etds-2578 Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification Youssef, Sarah Stock market manipulation detection is important for both investors and regulators. Being able to detect stock manipulation and preventing it gives investors the confidence in the market fairness and integrity. It also helps maintaining liquidity of the stocks and market efficiency. Implementing data mining algorithms in manipulation detection is a relatively recent technique but in the past few years there has been an increasing interest in it's applications in this domain. The benefit of monitoring manipulative trade behavior is that it can be implemented on live feed of stock data, which saves a lot of time in detecting stock price manipulation. This research implements machine learning algorithms in detecting trade manipulations where trade behaviors artificially impact the National Best Bid and Offer (NBBO) of traded stocks. Research methodology implemented is based on feature extraction using signal analysis, taking advantage of the similarity between physical signals measured by machines and raw financial data. Accordingly, Continuous Wavelet Transform (CWT) is applied on actual manipulation data for feature extraction, Principal Component Analysis (PCA) and factor analysis are used for dimensionality reduction and then Machine Learning Classifiers are trained and tested. Tick Bid/Ask Price and volume data of actual 15 manipulation cases published by the Security Exchange Center (SEC) was extracted from an online interface and labeled accordingly. This data was then used to train, and test 3 different classification models (XGBoost, KNN & SVM) and the outcome was compared accordingly. Results showed that introducing continuous wavelet transform enhances model accuracy, it increased precision results tremendously, while reducing recall values slightly. Adding PCA, reduced run time greatly, yet reduced the quality of some models prediction. Out of the three classifiers XGboost & KNN are showing the highest performance. 2021-06-15T07:00:00Z thesis application/pdf https://fount.aucegypt.edu/etds/1581 https://fount.aucegypt.edu/context/etds/article/2578/viewcontent/Sarah_Youssef_MSF_Thesis_Fall2020_Final.pdf Theses and Dissertations AUC Knowledge Fountain Machine Learning Classification Signal Processing Continuous Wavelet Transform Price Manipulation Layering Spoofing Business Analytics Business Intelligence Computational Engineering Finance and Financial Management Portfolio and Security Analysis |
| spellingShingle | Machine Learning Classification Signal Processing Continuous Wavelet Transform Price Manipulation Layering Spoofing Business Analytics Business Intelligence Computational Engineering Finance and Financial Management Portfolio and Security Analysis Youssef, Sarah Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title | Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title_full | Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title_fullStr | Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title_full_unstemmed | Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title_short | Stock Market Manipulation Detection Using Continuous Wavelet Transform & Machine Learning Classification |
| title_sort | stock market manipulation detection using continuous wavelet transform machine learning classification |
| topic | Machine Learning Classification Signal Processing Continuous Wavelet Transform Price Manipulation Layering Spoofing Business Analytics Business Intelligence Computational Engineering Finance and Financial Management Portfolio and Security Analysis |
| url | https://fount.aucegypt.edu/etds/1581 https://fount.aucegypt.edu/context/etds/article/2578/viewcontent/Sarah_Youssef_MSF_Thesis_Fall2020_Final.pdf |
| work_keys_str_mv | AT youssefsarah stockmarketmanipulationdetectionusingcontinuouswavelettransformmachinelearningclassification |