Improving Stock Market Close Price Prediction: A Generalized Linear Model Approach for NSE Banking Stocks

Stock market prediction is challenging due to its unpredictable and volatile nature. This research paper focuses on improving the precision of future stock price prediction of bank stocks on the National Stock Exchange (NSE) to generate substantial profits for investors. The Generalized Linear Model (GLM), with three independent variables in four combinations, is used to predict. For the study, ten stocks from NSE have been selected from 'finance.yahoo.com,' with five from each sector – private and public banks. We enhanced the model's performance by incorporating the high and low prices from the previous day to the previous day's closing price. For the study, we analyzed the stocks of banks, including Axis Bank Limited, Bank of Baroda, Bank of India, Canara Bank, City Union Bank Limited, The Federal Bank Limited, IndusInd Bank Limited, Indian Overseas Bank, Kotak Mahindra Bank Limited, and Union Bank of India. To assess the effectiveness of the model, we utilized several metrics, including Mean Square Error (MSE), Mean Squared Percentage Error (MSPE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The MSPE had improved by a minimum of 0.0001 and a maximum of 0.0028. The MAPE had improved by a minimum of 0.0002 and a maximum of 0.0554. The MSPE, MAPE, and R squared standard deviation showed that introducing additional independent variables boosted the model's performance. This combination of variables can be used.