Indian Journal of Science and Technology

Abstract

Objective: Predicting a smallcap company stock price with best accuracy using data science and machine learning techniques. Method: A Dataset of one year data collected from different sources like paytm money and yahoo finance. The prediction if a stock price involves like data collection, data pre processing, testing, training, fitting an algorithm and then prediction of a stock price finally find the best accuracy using machine learning techniques. Findings: This model get the accuracy of 98%. Lstm algorithm is giving the high accuracy rate than all other algorithms. We can predict the stock price using some other algorithms also. But till now the best accuracy rate given by LSTM algorithm only. Novelty: This study contributes to finding out the best algorithm to predict the stock price in real time market conditions. Sometimes we may not predict the stock price correctly due to international market conditions like war situations. In this case total index will show in negative only. This is the real challenge in future predictions of stock price. Ultimately predicting a stock price in all conditions and market situations is a great challenge till now. To make it success we have to introduce a new algorithm by making some changes in previous algorithms according to the recent trends markets.

Keywords: Numpy, Matplotlib, RNN, Machine learning, Small cap, Stock price