Name: Choi
Firstname: Jai Won
Title: statistician
Institution: National Center for Health Statistics
Street: 6525 Belcrest Road
City: Hyattsville, MD
Zip-Code: 20782
Country: U.S.A.
Phone: 301-458-4144
Fax: 301-458-4608
Email: jchoi@cdc.gov
Authors: Jai Won Choi and Balgobin Nandram National Center for Health Statistics, 6525 Belcrest Road, Rm. 915, Hyattsville, MD 20782
Title: Measure of Similarity between Two Objects with Common Variables
Abstract: This paper proposes a simple method how to measure the similarity between two objects, when they include one or more common variables. These variables can be continuous, discrete, or any type. After the variables are adjusted by their correction factor, we obtain one score from a pair of common variables. The weighted sum of these scores of all pairs can be used to measure the similarity between the two objects. This sum ranges from 0 to 1 where 1 suggests that the two objects are identical and 0 means that they are completely different. This is extended to the two groups, each including one or more objects. We obtain the group similarity score with the weighted average of the similarity scores from all possible pairs of objects from the two groups.
Key Words: Similarity score, Two groups, Matching variables, Weighed average.