A user friendly decision support system for dealing stocks using neural network
N. Baba, Masaki Morimoto, M. Naito, Toshiyuki Maeda, H. Matsuda · 2005
A user friendly decision support system for dealing stocks has been developed. Parameters in the neural network consisting of 15 input units, 2 hidden layers, and 2 output units are trained by the hybrid algorithm which combines the modified backpropagation method with the random optimization method. It is shown by several simulation results that the proposed decision support system is quite helpful for dealing stocks.