NSE Stock market prediction using Deep Recurrent Neural Network and comparison with ARIMA

Adithyan C Pankajakshan

Master's thesis

Abstract:

The National Stock Exchange (NSE) stock index is a crucial indicator of the performance of the Indian stock market and is used by investors and financial institutions to make informed decisions. Accurately predicting the future values of the NSE stock index is a challenging task. In this study, we compare the performance of two popular methods for predicting the NSE stock index: deep Recurrent Neural Networks (RNNs) and Autoregressive Integrated Moving Average (ARIMA) models. The results of our study show that the deep RNN model outperforms the ARIMA model in terms of prediction accuracy, with RNN having lesser RMSE value than ARIMA model. This study provides insights into the relative strengths and weaknesses of deep RNNs and ARIMA models for predicting the NSE stock index and highlights the potential for deep learning techniques to improve the accuracy of stock market predictions. The results of this study could be useful for investors, financial institutions, and researchers interested in stock market predictions and financial time-series analysis.

Keywords

Deep learning, Recurrent Neural Network, ARIMA

Language used: English

Date on which the thesis was submitted / produced: 1. 4. 2023

Thesis defence

• Date of defence: 22. 5. 2023
• Supervisor: doc. Ing. Arnošt Veselý, CSc.
• Reader: Richard Vágner, externi

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme:
Informatics