Used long short-term memory networks to conduct 3 experiments for predicting future prices of the NIFTY50 index using 20 technical indicators, 13 global indices and a combination of both with total of 33 features. The performance for all three was evaluated on MSE, MAE and R2 score.