Neutrosophic Interval Moving Average and Linear Regression Approach to Stock Market Volatility Prediction

Abstract

This study examines stock market closing prices using neutrosophic interval moving averages and neutrosophic interval linear regression techniques to address financial data uncertainty. The 5-day neutrosophic interval moving average achieved the lowest Mean Squared Error (MSE) of 43.43676, outperforming the 3-day (66.3552) and 10-day (89.1942) averages. In contrast, neutrosophic interval linear regression yielded a higher MSE of 126.7759, highlighting the superior accuracy of the 5-day moving average. This method successfully forecasted October values (25, 28, 29, 30), showcasing its reliability in trend analysis and prediction. The findings demonstrate the value of neutrosophic methods for enhanced stock market forecasting.