Official Statistics

Exploring the use of multiple imputation latent class modelling (MILC) for administrative data

A proof of concept project exploring how we might begin to apply multiple imputation latent class modelling to ethnicity administrative data and its potential to estimate classification error and produce estimates from linked, categorical administrative datasets that account for multiple sources of uncertainty.

Documents

Multiple imputation latent class modelling – an exploratory application to administrative data

Details

Official statistics are produced impartially and free from political influence.

Published 27 January 2023