|Friday, February 24|
|PS2 Poster Session 2 and Refreshments||
Fri, Feb 24, 5:15 PM - 6:30 PM
Conference Center AB
Pseudo-Maximum Likelihood Estimation with Sampling Weight for Modeling Count Data from a Complex Survey (303463)Mulugeta Gebregziabher, Medical University of South Carolina
*Lin Dai, Medical University of South Carolina
Keywords: zero-inflated, complex survey, pseudolikelihood, negative binomial
In this study we introduce a multilevel pseudo maximum likelihood method for weighted count data from a complex survey. We conduct a simulation study to assess the finite sample performance of this estimation method under several scaling methods of the survey sampling weights from the survey. In addition, different approaches for estimating the variance of parameter estimates corresponding to covariates that are included in the model are considered and compared in the simulation studies. The simulation results show that multilevel pseudo maximum likelihood estimations (MPML) with two-level sampling weight seems to provide better performance in terms of relative bias and 95% confidence interval coverage. We recommend MPML to analyze multilevel weighted count data. Among the four variance estimation approaches we considered, the empirical variance estimator provides values that reflect the true variance of the estimators. We use a real data example to illustrate the application of the best method to analyze count data from a two-stage survey study.