The data outlier elimination method based on grey relation

Some frequently used method of is data outlier elimination introduced, such as least square method and kalman filter method. On the base of this, A new method of data outlier elimination is proposed. It is based on grey relation and it proceeds the original data to find the data changing rule. Then it will make the regular data sequence at the same time it can establish the model of a differential equation. So it can eliminate the abnormal data and replace them with the prediction data. Through simulation, we can conclude that this method is feasible and it has the advantage in the field of processing the data outlier.