Systematic Review on Algorithmic Trading

Background: Algorithmic trading systems (ATS) are defined by the use of computational algorithms for automating financial transactions.They have become a critical part of modern financial markets because of their efficiency and ability to carry out complex strategies.Objective: This research involves a systematic review that assesses the market impact, technological advancements, strategic approaches and regulatory challenges related to algorithmic trading.Methods: Following PRISMA 2020 guidelines, this study conducts a systematic literature review by screening 1,567 articles across five academic databases, namely IEEE Xplore, ACM Digital Library, SpringerLink, Web of Science and SSRN.After applying predefined inclusion and exclusion criteria, 208 peer-reviewed journal and conference papers published between 2015 and 2024 are selected.The PICOC framework is used to define the review scope.Data are extracted using structured templates capturing study details, research objectives, artificial intelligence (AI) integration, profitability analysis and limitations.Tools such as Rayyan, NVivo, MS Excel and Zotero support the screening, coding and qualitative synthesis of findings.Results: AI methods, especially machine learning (used in 50% of the studies) and sentiment analysis (20%), significantly improve predictive accuracy and profitability.Most studies focus on equities (35%) and forex (30%), with high-frequency trading being the most examined strategy (30%).Challenges include latency (30%), scalability (25%) and regulatory issues (25%).Conclusion: Future research should prioritize ethical frameworks, regulatory clarity and wider access to AI-driven ATS components.This review provides a robust foundation for academics and practitioners to innovate and optimize algorithmic trading strategies.