Analysis of Financial Market Forecasting using Long Short-Term Memory (LSTM)

One of the most popular applications of machine learning is financial market forecasting. Share markets are dynamic, unpredictable, volatile and chaotic. Because of this, predicting stock prices is anything but simple. Understanding trends is very important to help investors in proper decision-making. In this research, we collected 1 year and 3 years data of National Stock Exchange of India (NSE) from Yahoo Finance and proposed a comprehensive customization of feature engineering. Deep learning-based models are used for predicting price trend of stock markets. The programming language used is Python. In this solution we have included pre-processing of the stock market dataset as well as utilized various feature engineering techniques and with a customized deep learning system is combined for the stock market price trend prediction. Open, High, Low and Close prices of stock are used as inputs to the model. The proposed model is applied on State Bank of India (SBI) and HDFC bank. Both shares are among NIFTY50. The models are evaluated using RMSE & MAPE. The low values of RMSE & MAPE show that the models are efficient in predicting stock price.