Sectoral indices like Bank Nifty are vital as they serve as the underlying for various investment instruments, including futures, options, mutual funds, and exchange-traded funds. Accurate forecasting of these indices is crucial for informed decision-making by investors, financial experts, and professionals. This study presents a comprehensive evaluation of three forecasting models – Random Walk (RW), Autoregressive Integrated Moving Average (ARIMA), and Artificial Neural Network (ANN) – in the context of India’s Bank Nifty index. The dataset encompasses daily adjusted closing prices spanning from January 2017 to June 2023, totalling 1596 observations. To facilitate the analysis, the dataset is divided into a training period (January 3, 2017, to May 3, 2023) and a testing period (May 4, 2023, to June 28, 2023). The study employs Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) as error metrics, consistently ranking the ARIMA model highest, followed closely by the ANN model, with the RW model performing less effectively. This research contributes to the field by comparing these models within the Bank Nifty index context and expanding the literature on forecasting sectoral stock market indices, particularly in emerging markets. The findings of this study have significant implications for academia, traders, investors, fund managers, and regulators.