Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{186098,
        author = {Naman Barola},
        title = {The use of Artificial Intelligence in Portfolio Management},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {115-121},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=186098},
        abstract = {Contemporary financial markets depend more on data and automated tactics, as a means of managing investment portfolios. This article investigates how some computational intelligence techniques – including basic ML and intricate models – are employed to implement thoroughly automated portfolio governance, notably in making independent purchase or sale judgments, and producing returns.

We inspect existing work, including examples of situations, to furnish realistic illustrations and data regarding portfolio returns where AI is used for regulation. Numerous kinds of predictive algorithms together with decision-making constructions get analyzed, like regression done the usual way, also forests generated via randomness, LSTM arrangements, and moreover RL representatives.

The outcome resulting from utilizing any algorithm undergoes testing by means of portfolio criteria utilized for measurement (e.g., accumulated gains, Sharpe quotient, greatest pullback) as contrasted and typical guidelines. Results originating within scholarly studies, plus company publications, get consolidated. As an instance, RL representatives of a specific kind displayed boosted Sharpe quotients than what mean-variance enhances did1, and groups involving trees crafted at arbitrary typically deliver shorter term forecasts of gains much better than only employing the serial aspects for single instants in time2,3.

Examining practical scenarios of accounts signals that funds based on computational smarts could achieve moderately increased earnings as measured across all revenues rather than competing monies but perhaps come short when viewed when altered to take risks in consideration4). It’s the final summary that AI mechanisms may augment several pieces of portfolio control, for example finding market data plus clocking trades, though it meets hurdles during use in genuine cases. By and large, proof gives the nod equally to assurance but limits that AI generates valuable portfolios by itself.},
        keywords = {},
        month = {October},
        }