MAXIMUM-LIKELIHOOD INFERENCE FROM SAMPLE SURVEY DATA

In this paper we present a general theory for maximum likelihood infer ence based on sample survey data. Our purpose is to identify and empha sise the recurring basic concepts that arise in the application of lik elihood methods, including the estimation of precision, to survey data . We discuss the problems generated by the effects of sample design, s election and response processes. We also discuss the problem of failur es of the model assumptions and the role of sample inclusion probabili ties in achieving robustness. We present two illustrative examples, on e of which illustrates the use of non-Gaussian models.