Neural Network with Partial Least Square Prediction Model Based on SSA - MGF

The primitive rainfall series be reconstructed and become as independent variables by singular spectrum analysis and mean generating function, so primitive rainfall series be as dependent variables. The factor affecting be withdrew by means of partial least squares method to extract the most important components so that it can be input as the neural network, and established the forecast model of the neural network with least squares regression based singular spectrum analysis and mean generating function, results show that the model is superior in predictions compared to the other models, and it is a useful model for the actual operational forecasting