Research Paper NUS Department of Statistics and Applied Probability Technical Reports 3/2010
A conditional empirical likelihood approach to combine sampling design and population level information
We consider methods to include sampling weights in an empirical likelihood based estimation procedure to augment population level information in sample-based statistical modelling. Our estimator uses conditional weights and is able to incorporate covariate information both through the weights and the usual estimating equations. We show that the estimates are strongly consistent, asymptotically unbiased and normally distributed. Moreover, they are more effcient than other methods. Our framework provides additional justication for inverse probability weighted score estimators in terms of conditional empirical likelihood. We give two applications to demographic hazard modelling by combining birth registration data with complex survey data to estimate annual birth probabilities.