A Brief Assessment on Stock Market Forecast using Long Short Term Memory (LSTM) Algorithm

The stock market price has increased significantly in the current period, attracting firm shareholders. Shareholders and investors both express a keen interest in stock market analysis and forecasting, which finally leads investors and other speculators to contribute to the company's financial...

Full description

Saved in:

Bibliographic Details
Published in:	2023 7th International Conference on Trends in Electronics and Informatics (ICOEI) pp. 1473 - 1478
Main Authors:	Kirubakaran, S Stewart, Selvadurai, J Jason, Ashok, V, Nivin, J Antony, Aron, A Dennies, Bijolin, E
Format:	Conference Proceeding
Language:	English
Published:	IEEE 11-04-2023
Subjects:	Heuristic algorithms Hyperparameter Tuning Long Short-term Memory Market Prediction Prediction algorithms Recurrent Neural Network Recurrent neural networks Root Mean Square Error SGD Optimizer Stock markets Task analysis Time series analysis
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The stock market price has increased significantly in the current period, attracting firm shareholders. Shareholders and investors both express a keen interest in stock market analysis and forecasting, which finally leads investors and other speculators to contribute to the company's financial success. A favorable prognosis could bring about significant advantages. In today's world, more improved models, various perspectives, and trend analysis tools are developed over time. Nevertheless, the most efficient analytical framework is Long Short-Term Memory, one of the algorithms of Recurrent Neural Networks. This algorithm can be used to produce precise results when the right parameters are used. To accomplish this, a dataset of stock market data must be compiled, and all stock closing prices must be measured using a variety of hidden layers and units. To improve accuracy, proposed work uses SGD optimizer and hyperparameter tuning. The algorithm is evaluated using root mean squared error. As a result of this methodology, historical datasets can be used to forecast the stock market more accurately.
DOI:	10.1109/ICOEI56765.2023.10125607