Stock Price Prediction Using Regression And Aritificial Neural Network

Computer Engineering

Stock markets are highly volatile and play a vital role in terms of contibution to the economy. Most stockbrokers use feedbacks within their community as well as trends of stocks to advice clients. This paper aims to explore the rereach and developement in stock market prediction applications implementations using regression analysis and artificial neural networks. The algorithm was implemented on a 210 days data of a particular stock and 30 days testing data. This system presents two approaches to analyze the data from stock market. One of them is regression analysis of data from widely available stock market records to predict future stock prices. The other approach is artificial neural network method. The proposed system will make use of neural networks with back propagation for parallel and distributed computing and regression analysis to prevent system failure in case of sudden fluctuation in the stock price due to uuprecedented events such as natural calamity,scams, etc . The use of back propagation ensures that error is reduced and accuracy is improved.The system does not guarantee a greater accuracy however, it certainly prevents the system from failing by using regression analysis for irregular inputs.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.