Effective market hypothesis using machine learning

Pilligundla, Niharika and Chimpiri, Yagnesh and Kandukoori, Uday Kiran and Jonnalagadda, Vaishnav Teja and Pagadala, Shiva Shashank (2025) Effective market hypothesis using machine learning. World Journal of Advanced Research and Reviews, 25 (2). pp. 553-560. ISSN 2581-9615

The Effective Market Hypothesis is designed to predict stock prices using Long Short Term Memory networks (LSTM), a kind of recurrent neural networks that are particularly suitable for modeling sequential and time series data. The LSTM model is designed to capture complex patterns, trends and potential market behaviours and, using historical stock price data and technical indicators, to model them, something that is often difficult in traditional statistical methods. Challenges such as market volatility, nonlinear dynamics and external factors that can have a significant impact on stock prices are also addressed. Model training and evaluation are then carried out rigorously, and the potential of LSTMs to increase the accuracy and reliability of stock price predictions is explored to deeper insights into market movements. In volatile market conditions, these findings are meant to help improve investment strategies, risk management practices, and the understanding of how machine learning techniques can be applied to financial forecasting.