A Review of Stock Price Prediction Based on LSTM and TCN Methods

Abstract

In financial analysis, stock price prediction is a difficult and important problem that has received a lot of attention from researchers and practitioners in recent years. The application of machine learning and artificial intelligence algorithms to stock price forecasting has demonstrated significant potential for increasing forecasting accuracy. Long momentary memory (LSTM) and transient convolutional networks (TCN) are two famous profound learning calculations that have been generally utilized for stock cost expectation. The most recent approaches to stock price prediction using LSTM and TCN methods are reviewed in this paper. We highlight the most recent research trends in this field and talk about these methods' benefits and drawbacks. Additionally, we discuss potential future research directions in this field. The survey is expected to give a knowledge into the present status of exploration on stock value forecast and guide specialists and experts in working on the exactness of expectation.