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{209092,
        author = {VENKATA MAHESWARA RAO ATCHULA and Dr. Pawan Kumar Pareek Phd},
        title = {Machine Learning-Assisted Validation of MACD Trading Signals Across Multiple Timeframes},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {5},
        pages = {704-711},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=209092},
        abstract = {Algorithmic trading systems based on conventional technical indicators can generate frequent signals during changing market conditions, including signals that may be difficult to distinguish from low-quality or short-lived opportunities. This study investigates a machine-learning-assisted validation framework for Moving Average Convergence Divergence (MACD) trading signals. The MACD strategy first generates candidate trading decisions using a conventional rule-based signal process. A machine-learning validation layer, implemented using XGBoost in the experimental framework, evaluates the candidate signal and applies a configurable confidence threshold before the trade is accepted. The study uses a seven-year BTCUSDT research dataset with four years used for training/development and three years used for out-of-sample testing. The experimental registry evaluates five timeframes—1-minute, 3-minute, 5-minute, 15-minute and 1-hour—and five ML confidence thresholds: 0.50, 0.53, 0.55, 0.57 and 0.60. The supplied results contain 25 ML test configurations and corresponding non-ML test configurations. Across the five timeframes, the ML configurations consistently reduce the number of executed orders and the reported maximum drawdown while producing modest increases in average win rate. The results therefore support interpreting ML primarily as a signal-selectivity and risk-control layer rather than as a universal profitability maximizer. The study provides an empirical framework for evaluating whether ML validation adds incremental value to a deterministic MACD trading strategy under multiple timeframes and confidence thresholds.},
        keywords = {MACD, Machine Learning, XGBoost, Algorithmic Trading, Signal Validation, BTCUSDT, Technical Analysis, Risk Management, Out-of-Sample Testing.},
        month = {October},
        }