Abstract: The Stock Market (SM) prediction price is one of an interesting field at present because it is a chaotic, non-linear, dynamic, non-stationary, noisy and quite difficult. Data mining has been effectively used in stock predicting hence researchers have explored Technical Indicators (TIs) to optimize the parameters. The main objective of this study is to predict NASDAQ stock index values market using TIs and develop method of multilayer perceptron neural networks based on an Optimization Model (OMLP) which aims to reduce the error factor depending on the Jacobian vector and Hessian matrix according to the convergence factor reach to zero. 10-fold cross validation and 70% holdout testing methods are used to train data. Further, precision, recall, F1 measure, specificity, accuracy, root mean squared error and mean absolute error are considered as evaluation criteria. Finally, a comparison has been implemented on input features with and without TIs. The results show using TIs satisfy better results of prediction. The accuracy rate is raised from 55.3% of standard features only to 79.97% of proposed approach which depends on TIs.