Imputation Method of Random Arbitrary Missing Data Based on Improved Close Degree of Grey Incidence

Through automatic identification and processing of the two sequence polylines positional relation in different time interval, the algorithm of close degree of grey incidence is improved, and a new missing data imputation method is proposed based on the improved absolute close degree of grey incidence. When the sample data exist a certain degree of random arbitrary missing, through calculating the absolute close distance and absolute close degree of grey incidence between different sequences, imputing the corresponding missing data with the attribute average of y non-missing sequence which the absolute close degree of grey incidence is highest, repeating iterations until the imputing data is converged, then getting the complete sample data set. The missing information is restored according to the relevance and proximity of the evaluated object, thereby improving the classification and imputing effect, and the feasibility and effectiveness of the algorithm are verified by an example and UCI data sets.