Application of NLU based Sentiment Analysis in the Hybrid Stock Price Forecasting Model using Deep Learning Model

For any developing market, like the Indian stock market, creating a stock market forecasting model is difficult. The reason being investor behavior on economic factors like changes in the western stock market, variations in the price of commodities on the international market, war-related escalations, and central bank actions influence the stock prices in emerging stock markets and less work on prior studies to consider investor behaviors based on several economic elements that affect price movement. To address this issue, a hybrid forecasting model is developed that uses Deep Learning algorithms through a multivariate Long Short-Term Memory (LSTM) model to predict future open and close prices by combining sentiment analysis data of economic information along with historical prices. The findings indicate that sentiment analysis data on economic information has an impact on stock price movement. The proposed model achieved better results in terms of coefficient for the determinant (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) and a Mean Absolute Percentage Error (MAPE) for the predicted result. The study findings can be used in future studies of other markets and asset classes where investor opinion toward certain events may have an impact on price fluctuations.