Stock Prediction using Gradient Descent Optimization

Abstract

Prediction stock market is the process of attempting to calculate the estimation of any company’s stock or other financial data. The proper prediction of a stock price can result in huge profits for the people investing and the various organizations. The focal point of this paper is to study and analyze various technologies that have been used to predict stock prices. We have examined various models such as Support Vector Machine (SVM), Multiple Regression, Neural Networks like Long Short Term Memory (LSTM) and Multilayer Perceptron (MLP), Ensemble learning. We wish to compare their efficiencies, analyze the time taken, resource used and labor compromised. We survey their demands, productivity and competence. We also wish propose another technique of Gradient Descent Optimization using RMSprops for predicting stock values.