Design Framework of Stock Price Forecasting Using Cascaded Machine Learning and Swarm Intelligence

Abstract

The variation of stock price declines the interest of investors and buyers in the stock market. The decline of interest creates a destructive impact on the stock market and national growth of the economy. The volatile nature of the stock market depends on the non-linear time series of the stock price. The stability of stock price is a challenging task to handle by conventional mathematical and neural network model. In this paper design cascaded machine learning algorithm for prediction of the stock price. The accurate and precise principle of cascading of machine learning improve the ratio of predication and decreases the rate of price variation. The primary issue of stock price variation is the strike price and risk factor of interest. The random behavior of risk factor, stick price and stock price carry on the unstable market situation. The variation of attribute behavior increased the value of data error. It degraded the strength of the market, for the depreciation of attribute randomness using swarm intelligence-based algorithm to select attribute of stock price and normalized the variation of data. The proposed algorithm simulated in MATLAB software and tested with a well-known dataset of Indian stock BSE. The design framework validates and compares with a machine learning algorithm.