Jaro-Winkler Distance

Stata ado file to calculate the Jaro-Winkler string distance between two strings. GitHub: https://github.com/jamesfeigenbaum/jarowinkler-ado

Use -ssc install jarowinkler- to install the ado file from Stata or visit http://ideas.repec.org/c/boc/bocode/s457850a.html

-jarowinkler- calculates the distance between two string variables using the Jaro-Winkler distance metric. The distance metric is often used in record linkage to compare first or last names in different sources. Jaro-Winkler modifies the standard Jaro distance metric by putting extra weight on string differences at the start of the strings to be compared. The metric is scaled between 0 (not similar at all) and 1 (exact match).

For more detail on the Jaro-Winkler method, see http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance and http://www.gabormelli.com/RKB/Jaro-Winkler_Distance_Function.

Jaro-Winkler implementation based on code from http://cs.anu.edu.au/~Peter.Christen/Febrl/febrl-0.4.01/stringcmp.py and https://github.com/miguelvps/c/blob/master/jarowinkler.c.