Estimating the Population Mean From a Simple Random Sample When Some Responses are Missing

Content Provider	CiteSeerX
Author	Lu, Jingsong Puleo, Elaine
Abstract	We develop a design-based prediction approach to estimate the finite population mean in a simple setting where some responses are missing. The approach is based on indicator sampling random variables that operate on labeled units (subjects). The missing data mechanism may depend on the subject, or on a selection (such as when the sample is assigned to different interviewers). The estimator (which equals the sample total divided by the expected sample size) emerges as the estimator using an approach that predicts the un-observed subject’s response using methods usually reserved for model-based inference. The methods are directly related to best linear unbiased predictors (BLUP) in finite population mixed models. When the probability of missing is estimated from the sample, the empirical estimator simplifies to the mean of the realized non-missing responses. The different missing data mechanisms are revealed by the notation that accounts for the labels and sample selections.
File Format	PDF
Access Restriction	Open
Subject Keyword	Simple Random Sample Different Interviewer Sample Selection Population Mean Non-missing Response Empirical Estimator Simplifies Finite Population Mixed Model Random Variable Un-observed Subject Response Design-based Prediction Approach Model-based Inference Simple Setting Linear Unbiased Predictor Finite Population Mean Sample Total Data Mechanism Expected Sample Size
Content Type	Text