This is a tutorial that Melody Jiang completed in 2019 as part of an independent study. This tutorial replicates the paper of Steorts, Hall, Fienberg (2016) and gives guidance to those who are wanting to study record linkage as an undergraduate or first year PhD student.
The tutorial does the following:
- Gives suggested background reading that one should complete before going through the tutorial
- Provides detailed derivations that may be helpful as one goes through those in the paper
- Provides code in R (as the code from the paper is fully in Java). Remark: the code does not provdide the split and merge step so the code will be quite slow as it utilizes a simple Gibbs sampler.