Dynamic Long Short-Term Memory Model for Stock Market Price Forecasting
Indrajit Sahu, Kiran Shankar Paira, Priti Rani Bhoi, Samrudhi Mohdiwale · 2024
This research paper aims to explore the utilization of Stacked Long Short-Term Memory (LSTM) method for prediction the stock market prices using machine learning. The objective of the research is to determine the performance of LSTM model in predicting stock prices and to compare its accuracy with other similar existing prediction methods. Utilizing Stacked LSTM entails the ability to layer numerous LSTM units on one another. Each LSTM layer possesses the capability to capture various levels of abstraction, thereby enhancing the model’s capacity to discern intricate patterns within the data. Historical stock data of NIFTY50 was collected for conducting this research, and an LSTM model was employed for the prediction of future prices of the stocks. The paper investigates different preprocessing techniques, such as data scaling and feature engineering. Data scaling is a preprocessing technique used to bring numerical features to a similar scale and Feature engineering associates building new features or transforming real features in the data to extract relevant information, such as creating technical indicators or incorporating time-based features. The LSTM model was successful to identify and capture fundamental trends and patterns of the NIFTY50 data, making it an efficient tool for stock price prediction. The work of the LSTM model for stock market price prediction was evaluated using several metrics, including accuracy, root mean squared error (RMSE), and mean absolute error (MAE). Accuracy was used as the primary metric for evaluating the performance of the model. The correctness of the model was decided by analyzing the predicted values with the real values in the testing dataset. Based on the findings, it is evident that employing an LSTM model can prove to be a valuable and effective approach for predicting stock prices. This can offer traders and investors valuable insights into market trends, aiding them in making well-informed decisions.