EUVIMEDEUVIMED entdecken
Crossref

Machine Learning-Driven Strategies for NASDAQ 100 Outperformance

Maxwell Fitchue, Kristin Stowe, Lisa Schwartz

Journal of Accounting and Finance · 19.06.2026 · 26 (2)

DOI: 10.33423/jaf.v26i2.8290 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Accounting and Finance, online erschienen am 19.06.2026. Verfasst von Maxwell Fitchue und 2 weiteren Autor:innen. Die Arbeit nennt 7 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This research evaluates an XGBoost-driven algorithmic strategy using macroeconomic indicators and technical filters to outperform the NASDAQ 100 benchmark. By testing "Magnificent 7" equities across 2018–2025 market regimes, the study highlights a critical trade-off: while VIX filters and hysteresis buffers successfully mitigate drawdowns during volatility, they create a lag that hinders returns during rapid recoveries. Additionally, while gradient boosting identifies complex non-linear relationships, it risks overfitting in stable markets. The findings suggest that adaptive hyperparameters are essential for balancing capital preservation with the momentum capture required for superior risk-adjusted returns.

Zitieren

APA

Fitchue, M., Stowe, K., & Schwartz, L. (2026). Machine Learning-Driven Strategies for NASDAQ 100 Outperformance. Journal of Accounting and Finance, 26(2). https://doi.org/10.33423/jaf.v26i2.8290

Vancouver

Fitchue M, Stowe K, Schwartz L. Machine Learning-Driven Strategies for NASDAQ 100 Outperformance. Journal of Accounting and Finance. 2026;26(2). doi:10.33423/jaf.v26i2.8290

BibTeX
@article{Fitchue2026,
  author = {Fitchue, Maxwell and Stowe, Kristin and Schwartz, Lisa},
  title = {Machine Learning-Driven Strategies for NASDAQ 100 Outperformance},
  journal = {Journal of Accounting and Finance},
  year = {2026},
  volume = {26},
  number = {2},
  doi = {10.33423/jaf.v26i2.8290}
}
Kostenlos starten →