A deep comprehensive model for stock price prediction

Due to many factors and noise involved in stock market, predicting stock prices is a challenging task. Existing approaches for stock price prediction use only a subset of features and factors that are effective in stock market. In this paper, we present a stock prediction model, which takes into account a diverse and comprehensive set of factors and elements that affect stock market. We accordingly present a comprehensive input model, as a generalization and improvement of existing input models, and feed it into a learning model to predict stock prices. Our learning model is a neural network, consisting of 4 hidden LSTM layers. We evaluate our proposed stock price prediction model over eight real-world datasets, and show its high performance and accuracy. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.