Predicting the Trends in Stock Movement Using Deep Neural Networks

Predicting the Trends in Stock Movement Using Deep Neural Networks

Priya Dohre1, Prof. Pankaj Rahguwanshi2

Abstract— Stock market forecasting is challenging and subject to uncertainties. Market dynamics can be influenced by a wide range of factors, including economic data, geopolitical events, investor sentiment, and unexpected events like natural disasters or health crises. Forecasting methods vary, ranging from technical analysis (studying historical price and volume data) to fundamental analysis (evaluating a company's financial health) to quantitative models and machine learning algorithms. Investors and analysts should approach stock market forecasts with caution, considering the limitations and potential biases of the methods used. Diversification, risk management, and a long-term perspective are often recommended to navigate the inherent uncertainty of financial markets. This paper presents a deep neural network model with data pre-processing for stock market forecasting. The results in terms of percentage error clearly indicate that the proposed approach performs better than existing work in the domain.

Keywords—Time Series Forecasting, Stock Trends, Neural Networks, Regularization, MAPE, Regression.