International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022
p-ISSN: 2395-0072
www.irjet.net
STOCK MARKET PREDICTION AND ANALYSIS USING MACHINE LEARNING ALGORITHMS Vasuki Rohilla1, Snehal Gore2, Akshara Garad3, Sumedh Dhengre4 4Professor,
Computer Dept. AISSMS College Of Engineering, Pune, Maharashtra, India 1,2,3 AISSMS College Of Engineering, Pune, Maharashtra. India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The goal of Stock Market Prediction is to forecast
about the stock being studied. However, numerous factors influence the stock market, including political events, economic conditions, and traders' expectations.
the future worth of a company's financial stocks. The nature of the stock market movement has always been ambiguous for investors because of various influential factors. This research aims to use machine learning and deep learning algorithms to reduce the risk of trend prediction considerably.
Researchers from a range of sectors, including computer science and business, are studying stock market projections. Researchers have experimented with a number of tactics and algorithms, as well as a mix of indications, to anticipate the market. The attribute that defines a prediction model is determined by factors that influence market performance.
Machine learning is a recent trend in stock market prediction technologies that provide projections based on the values of current stock market indices by training on their prior values. To develop accurate predictions, machine learning employs a range of models. The research focuses on stock value prediction using Linear regression, LSTM-based machine learning, and other ML models. There are several elements to examine, including open, close, low, high, and volume. The evaluation results will show that one of the models will outperform other prediction models for continuous data by a significant margin.
Time-series prediction is a commonly utilized technique in many real-world applications, including weather forecasting and financial market forecasting. It predicts the result in the following time unit using continuous data over a period of time. In practice, many time series prediction algorithms have proven to be effective. Recurrent Neural Networks (RNN) and their special types, Long-short Term Memory (LSTM) and Gated Recurrent Units, are now the most often used algorithms (GRU).
Key Words: LSTM, Linear Regression, Stock Market Indices, Recurrent Neural Network, Stacked LSTM.
This paper proposes to use LSTM, Linear Regression, and other ML models as ML tools for predicting the stock market prices for the next 30 days. Many people try to predict stock values, but it's a difficult task. Although perfect accuracy is unlikely, even simple linear models such as Linear Regression can be surprisingly close. Time series are used to represent stock trading data, and LSTM has the ability to learn extended observation sequences. This paper will also help us learn which machine learning algorithm will give the most accurate prediction. Long Short Term Memory (LSTM) networks are a type of recurrent neural network that can solve linear problems. A deep learning technique is LSTM. To learn very lengthy sequences, long-term memory (LSTM) units are required. The gated recurrent system in this form is more general. Because LSTMs address the evanescent gradient issue, they are more benign than other deep learning algorithms like RNN or classical feed-forward.
1. INTRODUCTION Financial markets are extremely volatile, and they create enormous volumes of data on a regular basis. It is the most widely traded financial instrument, and its value fluctuates rapidly. Stock prices are forecasted to determine the worth of a company's stock or other financial instruments traded on stock exchanges in the future. The stock market allows investors to purchase shares of publicly traded corporations through exchange or over-the-counter trading. This market has provided investors with the opportunity to make money and live a prosperous life by investing small quantities of money at low risk compared to the risk of starting a new business or the necessity for a high-paying job. Many factors influence stock markets, resulting in market uncertainty and excessive volatility. Although humans can take orders and transmit them to the market, automated trading systems (ATS) run by computer programs can submit orders faster and more accurately than humans. However, implementing risk strategies and safety measures based on human judgments is essential to evaluate and control the performance of ATSs. When developing an ATS, many factors are taken into accounts, such as the trading strategy to be used, complex mathematical functions that reflect the state of a specific stock, machine learning algorithms that allow for the prediction of future stock value, and specific news
© 2022, IRJET
|
Impact Factor value: 7.529
2. LITERATURE SURVEY A blindfolded monkey throwing darts at a newspaper stock listing should do as well as any investment professional, according to Princeton University economist Burton Malkiel, who argues in his 1973 book, If the market is genuinely efficient, and a share price reflects all aspects as soon as they are made public, a blindfolded monkey tossing darts at a newspaper stock listing should do as well as any investing
|
ISO 9001:2008 Certified Journal
|
Page 1977