Introduction

Stocks Prices are a fascinating thing for us. We invest in Stocks of companies which we think will increase in value, thereby making us wealtheir. Generally, the Stock Prices of well performing companies go up gradually making us want to invest in these companies. 

One of the intriguing things about Stock Prices is that it is difficult to predict their future prices. They change based on a many factors and for lay persons like us, it is very difficult to predict their change over time. Think of a situation that we know if the price of Reliance Industries will increase for the next one week. If we knew, we will put our money before it increases. But we don’t know that. 

One of the features of Stock Prices is that is they are Time Series data, i.e. they change over time, they change everyday. The price movement also follow patterns most of the time. 

In this project, we will consider and treat Stock price as a Time Series data with statistical patterns underneath. With that assumption, we will train a Time Series model using the ARIMA Algorithm and use that model to predict future stock prices. 

Note of caution: In reality stock price prediction used by financial organisations use hundreds of different parameters other than daily price data. Hence, please do not assume your project Model will be good enough to predict actual Stock Prices. This is a learning project to understand how to create Time Series Models using Machine Learning algorithms and predict future values of the Series.