Predicting Stock Price in Python Using TensorFlow and Keras

International Journal of Research and Scientific Innovation (IJRSI) | Volume VIII, Issue VI, June 2021 | ISSN 2321–2705

Predicting Stock Price in Python Using TensorFlow and Keras

Orlunwo Placida Orochi, and Ledesi Kabari
Computer Science Department, Ignatius Ajuru University of Education

Abstract: One of the most important practices in the financial world is stock trading. The act of attempting to forecast the future value of a stock or other financial instrument listed on a stock exchange is known as stock market prediction. This paper discusses how Machine Learning can be used to predict a stock’s price. When it comes to stock forecasts, most stockbrokers use technical and fundamental analysis, as well as time series analysis. Python is the programming language used to forecast the stock market. In this paper, we propose a Machine Learning (ML) method that will be trained using publicly accessible stock data to obtain intelligence, and then use that intelligence to make an accurate prediction. In this context, this research builds a neural network in TensorFlow and Keras that predicts stock market, which is basically a Python scraper that extracts finance data from the Yahoo Finance platform; more precisely, a Recurrent Neural Network with LSTM cells was constructed, which is the current state-of-the-art in time series forecasting.

Keywords: Stock, Artificial Neural Network, RNN, LSTM, Machine Learning, Prediction, Tensorflow, Keras, Artificial Intelligence.

I.INTRODUCTION

The financial market is a complex, composite mechanism that enables people via virtual broker sponsored platforms to buy and sell currencies, stocks and equities and derivatives. The stock market allows investors, through either exchange or over-the-counter trading, to own shares in the public company. This market gives investors the opportunity to make a living by invests in small initial sums of capital, a low risk compared with the risk of opening a new company or a high level of wage proficiency. Many factors that create uncertainty and high volatility in the market are affecting stock markets. However, macro or microeconomics like interest rates, exchange rates and monetary policy can easily influence inventory price, which makes prediction into a difficult task. Researchers and speculators based their research on stock market prediction over decades based on great profit in stock market investment. The prediction of the stock price movement was predicted by the traditional statistical methods such as logistic regression, the exponential average, ARIMA, and GARCH [9].

Read More