convolutional neural network (cnn)

Non-linearity feature and high fluctuations in financial time series have made the forecasting of stock prices and financial indicators face many challenges. However, recent developments in deep learning (DL) models with structures such as long-short-term memory (LSTM) and convolutional neural network (CNN) have made improvements in the analysis of this type of data. Another approach that can be effective in the analysis of financial time series is the decomposition of capital market signals through algorithms such as complete integrated empirical mode decomposition (CEEMD). Considering the importance of forecasting in the financial markets, in this research, by combining deep learning models and complete integrated empirical mode decomposition (CEEMD), The hybrid CEEMD-DL(LSTM) model has been used to forecast the Tehran Stock Exchange index. In this regard, the daily data of the total index of the Tehran Stock Exchange in the period of 2012/12/01 – 2022/02/20 be used and the results were compared with the results of competing models based on efficiency measurement criteria. Based on the obtained results, the introduced model (CEEMD-DL(LSTM)) has higher efficiency and accuracy in stock exchange index forecasting. Accordingly, the use of this model in financial forecasts is suggested.

Algorithms based on a Convolutional Neural Network (CNN), which is a branch of Deep Learning (DL), have seen significant progress in picture and video analyses in recent years. Success of these new models has led to widespread use of them in various fields, including text mining and time series data. DL is part of a broader family of machine learning methods that attempts to model high-level concepts using learning at multiple levels and layers and extract features of higher levels from the raw input. This survey investigated the abilities of different CNN architectures to predict stock prices. Upon running the model with various architectures and parameters for the stock price of Esfahan Steel Company, the results showed that a CNN with max-pooling layers (a combination of Batch size=64, filters=256, and ReLU Activation Function) and Mean Absolute Percentage Error (MAPE) of 1.79% and Normalized Root Mean Square Error (NRMSE) of 2.71% had a higher prediction accuracy than other CNN architectures and Recurrent Neural Network (RNN).