Prediction of stock price growth for novel greedy heuristic optimized multi-instances quantitative (NGHOMQ)

摘要

Abstract A large volume of data is constantly accumulating in contemporary information technology relating to sophisticated computing. Stock price prediction is a popular topic and a difficult task due to the randomness and complexity of stock market-related computing systems. As stock market information web content develops, researchers and investors typically extract different indicator aspects, such as attitudes and events, from stock market real-time financial data predictions. Because of the current financial and unknown aspects in the stock market arena, stock price prediction is a difficult task, even though conventional authors worked on neural networks to improve stock price prediction in several financial domains. In this research, I present a Novel Greedy Heuristic Optimized Multi-instance Quantitative (NGHOMQ) strategy to investigate needed data from factors while rejecting their parameter interactions to improve index-based composite stock market movement prediction in multi-instance quantitative data. By employing unique heuristic calculations to describe sequential stock price-related events, it may be used to blend attitudes and events as well as to analyze quantitative data in a comprehensive way. In heuristic modes with numerous instances, Pareto optimization is utilized to prevent stock price prediction based on optimal statistical performance. Furthermore, our proposed approach may detect data input to produce stock market price forecasts in financial computing technologies. The efficiency of NGHOMQ compared to traditional neural network related frameworks and approaches is demonstrated using data from the Indian stock market.