Predicting Indian Stock Prices by Using Sentiment Analysis and Natural Language Processing

Abstract

This study employs traditional empirical models to investigate the impact of sentiments on financial market volatility through the use of financial indicators. In this paper, we apply recent methods for text-based sentiment analysis of the market from relevant news articles about the financial and business markets. Two separate market sentiments—positive and negative sentiments—are created from various emotions that can be identified using standard Natural Language Processing (NLP) techniques. The research also makes an effort to apply the previously mentioned market sentiments to a refined iteration of the asymmetric GARCH model of conditional volatility for the Indian stock market (Sensex) for the period from January 1, 2011, to December 31, 2022. Empirical findings indicate that negative market sentiment predominates over positive ones and that noise trading in financially undeveloped Indian stock market.