Trend-encoded Probabilistic Multi-order Model: A Non-Machine Learning Approach for Enhanced Stock Market Forecasts

TL;DR

This paper introduces TeMoP, a probabilistic multi-order model that uses trend encoding to improve stock market forecasts, demonstrating superior accuracy and robustness compared to machine learning models across multiple international datasets.

Contribution

The study presents a novel trend-encoded probabilistic model for stock prediction that outperforms existing machine learning approaches in accuracy and stability.

Findings

TeMoP outperforms machine learning models in prediction accuracy.

TeMoP demonstrates robustness across diverse stock datasets.

The model effectively captures multi-order temporal dependencies.

Abstract

In recent years, the dominance of machine learning in stock market forecasting has been evident. While these models have shown decreasing prediction errors, their robustness across different datasets has been a concern. A successful stock market prediction model minimizes prediction errors and showcases robustness across various data sets, indicating superior forecasting performance. This study introduces a novel multiple lag order probabilistic model based on trend encoding (TeMoP) that enhances stock market predictions through a probabilistic approach. Results across different stock indexes from nine countries demonstrate that the TeMoP outperforms the state-of-the-art machine learning models in predicting accuracy and stabilization.