The Optimization of GM (1, N) Model and New Method of Forecasting

Abstract

On one hand, reconstruct a new GM (1, N) model equation by using a kind of optimized background value, and we discover that the new GM (1, N) model has higher simulated value and precision obviously, especially the series of data change sharply. On the other hand, deduce a forecasting formula that can suit the situation which the drive coefficient matrix of GM (1, N) model equation group is reversible, overcome the defect of nested model which require the drive coefficient matrix of GM (1, N) model equation group should be triangular array. Using the new forecasting formula, we can discover the new GM (1, N) model has higher prognostic precision.