IoT Monitoring Stock Price Forecasting by Using Machine Learning Techniques

Abstract

Time series forecasting has been widely used to determine the future prices of stocks, and the analysis and modeling of finance time series importantly guide an investor’s decisions and trades. In addition, in a dynamic environment such as the stock market, the non-linearity of the time series is pronounced, immediately affecting the efficacy of stock price forecasts. Thus, this work proposes an intelligent time series prediction system that uses a machine learning technique system optimized by PSO metaheuristic optimization for the purpose of predicting stock prices one-step ahead. It may be of great interest to investors who do not possess sufficient knowledge to invest in companies in different fields. In this paper, the prediction results are monitored on real-time by users based on an IoT platform as Thing Speak. After our predicted indicator is calculated on MATLAB environment, they're sent to Thing Speak platform in order to synthesis and notify to the non-professional users the future stockprice and suggest their trading actions via email without delays.