An optimization method of estimating parameters in GM (1, 1) model

According to Principle of New Information Priority in grey theory, giving the new information larger weight in modeling can improve the effectiveness of gray model. For the GM (1, 1)model has very small samples, and the high overall simulation accuracy does not necessarily guarantee high prediction accuracy, we put forward weighted least square method to estimate parameters in GM (1, 1) model. Focusing on improving the simulation accuracy of new information, focusing on grasping the latest development´ law of things, and aiming at improving the prediction accuracy by giving residual sum of squares of new information larger weight. Finally, we use an example to verify the practicality and reliability of the model.