Search

Missing data are common in many experiments, including surveys, clinical trials, epidemiological studies, and environmental studies. Unconstrained likelihood inferences for generalized linear models (GLMs) with nonignorable missing covariates have been studied extensively in the literature. However, parameter orderings or constraints may occur naturally in practice, and thus the efficiency of a statistical method may be improved by incorporating parameter constraints into the likelihood function. In this paper, we consider constrained inference for analysing GLMs with nonignorable missing covariates under linear inequality constraints on the model parameters. Specifically, constrained maximum likelihood (ML) estimation is based on the gradient projection expectation maximization approach. Further, we investigate the asymptotic null distribution of the constrained likelihood ratio test (LRT). Simulations study the empirical properties of the constrained ML estimators and LRTs, which demonstrate improved precision of these constrained techniques. An application to contaminant levels in an environmental study is also presented. [ABSTRACT FROM AUTHOR]

Human biomonitoring studies in the Canadian Arctic have measured a wide range of metals and persistent organic pollutants in Aboriginal and non-Aboriginal mothers during two time periods in the Northwest Territories and Nunavut. This analysis provides preliminary estimates on sample sizes and sampling frequencies required to measure significant changes in maternal blood concentrations for PCB 153 and total mercury. For example, sample sizes of 35-40 mothers permit the detection of a 40% decrease in these chemical concentrations between two groups (e.g. communities or regions). Improvements in method sensitivity can be achieved by on-going sampling over multiple time periods (e.g. 4 or 5) in these regions, or increasing sample sizes. [ABSTRACT FROM AUTHOR]