One of the major difficulties in business and economics is to manage data accurately. Currently, approaches in stock forecasting have been rough set theory, time series and artificial neural networks. One of the most popular types of neural network has been the back propagation Neural Network. However, there are still problems with these models and there is a hidden layer. New models-Partially Connected Neural Evolutionary model (PARCONE) can correct this flaw. Parcone for the discovery of the more hidden knowledge stored within the historic time series data are needed in order to deal with the timely trading problem, i.e., buy, hold or sell within a time constraint.