StockForecastGenie Pro: A Volatility-Aware Machine Learning Decision-Support System for Short-Horizon Equity Index Forecasting

Published in: Proceedings of the 7th National HBCU Blockchain, FinTech, and AI Conference
Conference: 7th National HBCU Blockchain, FinTech, and AI Conference
Conference Location: Nashville, Tennessee
Date of Conference: November 8–10, 2026
Publisher: National Center for the Study of Blockchain and FinTech, Morgan State University

Published: 2026
Publication Type: Full Paper
Pages: 83-97
Paper ID: 5

Author(s)

Chibuike Chiedozie Ibebuchi
Department of Mathematics, Morgan State University

Abstract

Forecasting short-term movements in financial markets remains challenging because market prices are influenced by rapidly changing economic conditions, investor sentiment, and volatility. While machine-learning methods have demonstrated promise for financial forecasting, many forecasting tools remain difficult for non-technical users to interpret and often provide predictions without contextual explanations. This paper presents StockForecastGenie Pro, a volatility-aware machine-learning decision-support system designed to forecast short-horizon movements in major United States equity indices while providing transparent and interpretable signals. The system forecasts future index levels, point gains, and percentage returns for the S&P 500 and Dow Jones Industrial Average using a stacked ensemble of machine-learning models. Forecast outputs are combined with market-volatility information derived from the CBOE V olatility Index (VIX) to generate plain-language signals ranging from Strong Negative to Strong Positive. The platform also incorporates confidence estimates, simple benchmark comparisons, and broader market context to support user interpretation. A rolling historical evaluation was conducted, and results indicate that forecasting performance improves substantially as forecast horizon increases. At the five-day horizon, the stacked ensemble achieved directional accuracy of 73.2% for the Dow Jones Industrial Average and 70.9% for the S&P 500. Compared with a no-change benchmark, the system reduced forecast error by approximately 24.3% for the Dow Jones Industrial Average and 22.4% for the S&P 500. The results demonstrate that machine-learning ensembles can provide meaningful improvements over simple forecasting approaches while maintaining interpretability for research, educational, and decision-support applications.

Attachment: View Full PDF