Forecasting Bank Nifty Close Prices: An Evaluation of Time Series and Hybrid Models

In India’s economy, the stock market plays a critical role. Making accurate stock price predictions is crucial for both individuals and corporations, since it helps in guiding investment decisions and managing risks effectively. This study focuses on forecasting the close price of Bank Nifty, which is an index that includes banking stocks listed on the National Stock Exchange(NSE). Multiple time series models such as Autoregressive Integrated Moving Average(ARIMA), Gated Recurrent Unit(GRU) and Prophet, also their different hybrid models are assessed for their predictive capabilities using metrics like Mean Absolute Percentage Error(MAPE), Root Mean Squared Logarithmic Error(RMSLE) and Mean Squared Logarithmic Error(MSLE). The study also highlights the efficiency of Auto ARIMA. Emphasizes that the GRUPROPHET model stands out as the precise predictor with the ARIMA-GRU model coming in a close second. The study underscores using the GRU-PROPHET model for predicting the closing price of Bank Nifty. Stresses the significance of employing advanced forecasting methods, in financial markets.