A Modified Neural-Network–Based GM(1,1)

Inspired by neural-network-based GM(1,1) (NN-GM(1,1)), this study attempts to drive a more simple and efficient learning rule to enhance the fitting/forecasting ability and convergence speed of the grey neural network. Simulation results on two real datasets show that the proposed modified NN-GM(I,I) performs better than the original NN-GM(I,I) in terms of fitting and forecasting accuracies. In addition, the modified NN-GM(I,I) has faster convergence speed.