Analysis of Stock Market Predication Based on Machine Learning Algorithm

Abstract

Prediction of the stock market decides the financial future of traders and investors. The volatile nature of stock market trends decreases the financial investment in the market. For the prediction of the stock price, various parametric and non-parametric computational models are used. The non-parametric models are conventional approaches and have bottleneck problems regarding the predication of a stock price. The non-parametric model enriches the prediction capacity of stock price trends. The process of non-parametric models used the methodology of artificial neural network and machine learning algorithms. The primary variation of the stock price depends on the random nature of the stock price attribute. Swarm based algorithms reduce the variation factor of a stock price. This paper study the stock market prediction based on machine learning algorithms and swarm-based algorithms. The validation of these algorithms used stock market data and measured standard parameters of variation—these algorithms are implemented in MATLAB software and test bench results