Using Transformer Models for Stock Market Anomaly Detection

Anomaly detection is an important task in financial markets. Detecting anomalies is difficult due to their rarity, multitude of parameters, and lack of labeled data for supervised learning models. Additionally, time series data used in financial models present unique challenges such as irregul...

Full description

Bibliographic Details
Main Authors: Biriukova, Kseniia, Bhattacherjee, Anol
Format: Article
Language:English
Published: INTI International University 2023
Subjects:
Online Access:http://eprints.intimal.edu.my/1848/
http://eprints.intimal.edu.my/1848/1/jods2023_21.pdf
_version_ 1848766851795910656
author Biriukova, Kseniia
Bhattacherjee, Anol
author_facet Biriukova, Kseniia
Bhattacherjee, Anol
author_sort Biriukova, Kseniia
building INTI Institutional Repository
collection Online Access
description Anomaly detection is an important task in financial markets. Detecting anomalies is difficult due to their rarity, multitude of parameters, and lack of labeled data for supervised learning models. Additionally, time series data used in financial models present unique challenges such as irregularity, seasonality, changing trends, and periodicity in data. While prior anomaly detection approaches have used ARIMA and LSTM models, in this paper, we employ a new transformer�based model called TranAD to compare stock market data with its predicted version, measuring deviations from normal price data for anomaly detection. We find that TranAD is an effective approach for financial anomaly detection with a high level of accuracy. We expect that this research will contribute to better detection of financial anomalies and improve market surveillance
first_indexed 2025-11-14T11:57:43Z
format Article
id intimal-1848
institution INTI International University
institution_category Local University
language English
last_indexed 2025-11-14T11:57:43Z
publishDate 2023
publisher INTI International University
recordtype eprints
repository_type Digital Repository
spelling intimal-18482023-12-05T03:54:08Z http://eprints.intimal.edu.my/1848/ Using Transformer Models for Stock Market Anomaly Detection Biriukova, Kseniia Bhattacherjee, Anol Q Science (General) QA76 Computer software Anomaly detection is an important task in financial markets. Detecting anomalies is difficult due to their rarity, multitude of parameters, and lack of labeled data for supervised learning models. Additionally, time series data used in financial models present unique challenges such as irregularity, seasonality, changing trends, and periodicity in data. While prior anomaly detection approaches have used ARIMA and LSTM models, in this paper, we employ a new transformer�based model called TranAD to compare stock market data with its predicted version, measuring deviations from normal price data for anomaly detection. We find that TranAD is an effective approach for financial anomaly detection with a high level of accuracy. We expect that this research will contribute to better detection of financial anomalies and improve market surveillance INTI International University 2023-12 Article PeerReviewed text en cc_by_4 http://eprints.intimal.edu.my/1848/1/jods2023_21.pdf Biriukova, Kseniia and Bhattacherjee, Anol (2023) Using Transformer Models for Stock Market Anomaly Detection. Journal of Data Science, 2023 (21). pp. 1-8. ISSN 2805-5160 http://ipublishing.intimal.edu.my/jods.html
spellingShingle Q Science (General)
QA76 Computer software
Biriukova, Kseniia
Bhattacherjee, Anol
Using Transformer Models for Stock Market Anomaly Detection
title Using Transformer Models for Stock Market Anomaly Detection
title_full Using Transformer Models for Stock Market Anomaly Detection
title_fullStr Using Transformer Models for Stock Market Anomaly Detection
title_full_unstemmed Using Transformer Models for Stock Market Anomaly Detection
title_short Using Transformer Models for Stock Market Anomaly Detection
title_sort using transformer models for stock market anomaly detection
topic Q Science (General)
QA76 Computer software
url http://eprints.intimal.edu.my/1848/
http://eprints.intimal.edu.my/1848/
http://eprints.intimal.edu.my/1848/1/jods2023_21.pdf