data one;
input ID_A ID_B Match_score;
cards;
123 777 28.1
124 778 15.6
125 787 19.7
125 799 18.9
126 762 36.1
127 762 55.1
127 777 28.7
128 999 19.5
129 781 18.2
; I just performed probabilistic linkage on two datasets. The output dataset called "one", contains the identification number from both original datasets, ID_A, the other ID_B, with a linkage score "match_score". There are numerous combinations of ID_A and ID_B. I want to select only the top linkage to pair then remove them from the selection process for further linkages. An ideal output would be... ID_A ID_B Match_score 127 762 55.1 123 777 28.1 125 787 19.7 128 999 19.5 129 781 18.2 124 778 15.6 ID_A: 126 wouldn't match because of the ID_B (762), match_score is higher for another ID_A (127). ID_B: 799 wouldn't match because ID_A(125) had a larger match_score with (787) Any help would be greatly appreciated! Edited 28.7 to 28.1...thanks for pointing out
... View more