This study examines applicability of Nonlinear autoregressive neural network (NARNET) for prediction of intraday stock of S&P Bombay Stock Exchange (BSE) Sensex-50 index. NARNET is widely used for multistep time-series forecasting. The model has been trained using three different training algorithms i.e. Levenberg-Marquardt (LM), One-Step Secant (OSS) and Broyden–Fletcher–Goldfarb–Shanno-Quasi Newton (BFGS-QN). The current work explores different training algorithm for time-series prediction of Intraday stocks for a duration of 10-min considering training data of past 1 h. The performance of proposed algorithms has been monitored in terms in terms of mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). The results clearly highlight superior performance of LM-algorithm giving much lower values of error responses compared to other training algorithms.