Evaluation of small area estimation techniques for the Canadian Labour Force Survey

Estimates from sample surveys are sometimes required for domains whose boundaries do not coincide with those of design strata. Taking the Canadian Labour Force Survey as an example of a survey utilizing a clustered sample design, some alternative small area estimation techniques available in the literature are evaluated empirically including synthetic, domain (simple and post-stratified) and composite estimators which are linear combinations of synthetic and post-stratified domain estimators. A sample dependent estimator which attaches weight to the post-stratified domain estimate depending on the amount of sample in the domain is proposed and its performance is also evaluated.