Calculates Felligi-Sunter weights and posterior zeta probabilities for matching patterns observed in a larger population that are not present in a sub-sample used to estimate the EM.
emlinkRS(patterns.out, em.out, nobs.a, nobs.b)emlinkMARmov returns a list with the following components:
The posterior match probabilities for each unique pattern.
The posterior probability of a pair matching.
The posterior probability of a pair not matching.
The posterior of the matching probability for a specific matching field.
The posterior of the non-matching probability for a specific matching field.
The posterior probability that a pair is in the matched set given a particular agreement pattern.
The posterior probability that a pair is in the unmatched set given a particular agreement pattern.
Counts of the agreement patterns observed, along with the Felligi-Sunter Weights.
The number of iterations it took the EM algorithm to converge.
The number of observations in dataset A.
The number of observations in dataset B.
The output from `tableCounts()` or `emlinkMARmov()` (run on full dataset), containing all observed matching patterns in the full sample and the number of times that pattern is observed.
The output from `emlinkMARmov()`, an EM object estimated on a smaller random sample to apply to counts from a larger sample
Total number of observations in dataset A
Total number of observations in dataset B
Ted Enamorado <ted.enamorado@gmail.com> and Ben Fifield <benfifield@gmail.com>