METHOD FOR COMPOSING ONE FILE WITH THE STATISTICAL MATCHING USING DOUBLE REGRESSION TREE MODEL FROM TWO FILES

摘要

A file matching method for composing one file through statistical matching by using a double regression tree model in two files is provided to obtain one complete file from two different files extracted from the same population by matching two different files statistically, and reduce cost/time needed for collecting data to be obtained from two different files. Different files respectively including a common variable and a consecutive unique variable having a different value are extracted as population. The common variable is set as an input variable and the consecutive unique variable is set as a target variable in recipient and donor files of the extracted different files. The recipient file is set as training data and the donor file is set as verification data. One complete file is generated by adding the unique variable(240) included in the donor file to an empty area of the unique variable(140) assigned to the recipient file based on the common variable(120) through statistical matching by using a double regression tree model when the common variables of the different files is identical with each other.