A Novel CNN-LSTM Model for Predicting the Railway Sector Stock Price in the Tehran Stock Exchange

Abstract

As a country's financing arm, the capital market can play a crucial role in long-term projects and be responsible for large-scale financing. While the development of rail transport has exceptionally positive and significant effects on the country's macroeconomy, and investment in this sector leads to the prosperity of other economic sectors, the return on investment may be delayed or unclear. Consequently, this uncertainty diminishes the investment appeal in this field. Thus, predicting the price and future value of railway company stocks on the stock market is vital due to the need for the foresight to take safe investment measures and achieve sustainable financing. To this end, this paper investigated the use of deep learning methods to forecast the closing price of MAPNA and Toucaril stocks on the Tehran Stock Exchange. Stock prices were predicted using deep neural networks, including Long Short-Term Memory (LSTM), One-dimensional Convolutional Neural Networks (1D-CNN), and CNN-LSTM networks as a hybrid model. Finally, to evaluate the performance of the models, MAE, MSE, RMSE, MAPE, and R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> were used as evaluation criteria. The results indicate that deep learning models can forecast stock prices with accuracy. In this study, CNN-LSTM neural network produced the best results.