Stock Price Prediction Using Machine Learning

EasyChair Preprint 7022

Abstract

Several recent studies have made the use of machine learning in the field of quantitative finance, investment process, as well as, predicting prices of managing and constricting entire portfolio of assets and many other operations that can be covered by machine learning algorithms. In layman's terms, machine learning is used for all algorithmic methods using computers to reveal patterns based only on the data and without programming instructions. For special asset selections and quantitative finance, several machine learning models provide an extensive array of methods that can be used with machine learning to forecast the required future assets value. These types of models offer a mechanism that can combine weak sources of information, making it a strange tool that can be used efficiently. 
Neural network studies were originally started to map the human brain and understand how humans make decisions but algorithm tries to remove human emotions altogether from the trading aspect. We sometimes fail to realize that the human brain is quite possibly the most complex machine in this world and has been known to be quite effective at coming to conclusions in record time.

Predicting the future movement of security is the center of the industry of quantitative trading, as the future trading strategy is deployed and created based on our view of the financial market in the future. The trading area has two different methods, namely fundamental analysis, and quantitative trading. 
In this project, we have compared some of the existing neural networks (SimpleRnn, GRU, LSTM) and selected the neural network in which we were able to obtain maximum accuracy in quantitative trading.

Keyphrases: GRU, LSTM, SimpleRNN, Stock price prediction, machine learning, neural networks

@booklet{EasyChair:7022,
  author    = {Aashay Chaudhari},
  title     = {Stock Price Prediction Using Machine Learning},
  howpublished = {EasyChair Preprint 7022},
  year      = {EasyChair, 2021}}