Copyright © 2026 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{207479,
        author = {S. ANU and Dr. S. RITA},
        title = {A Comparative Study of Manual Time Series Forecasting and AI Tools for Silver Price Prediction Using Agent AI},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {3},
        pages = {1276-1284},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=207479},
        abstract = {Predicting silver prices matters greatly - it affects decisions made by investors, policy makers, and finance experts because silver functions both in industry and as an asset. One reason this research exists? To weigh classic time-series approaches against modern artificial intelligence methods when estimating future values of silver. Past market figures form the backbone of several predictive systems tested here: LSTM networks take shape alongside Auto ARIMA setups, Seasonal Naive patterns emerge, while AutoGluon powers fully automatic machine learning versions. Another layer appears through inclusion of a third-party AI predictor brought in purely for contrast purposes.
Despite its simplicity, time series prediction often relies on established methods such as the Wilder approach when estimating upcoming values. Performance gets measured through common indicators including MAE, RMSE, and MAPE - metrics widely accepted across evaluation practices. Instead of manual comparison, a system powered by Agent AI handles scoring and ordering of different models without human bias creeping in. Findings show Auto ARIMA achieving better accuracy than both deep learning setups and those built with AutoML tools on this particular data set. What stands out is how traditional statistical techniques still hold strong under real testing conditions, especially when matched against more complex alternatives.},
        keywords = {},
        month = {August},
        }