Statistics Colloquium
September 16 (Fri) 11 a.m.
Alavi Commons Room, 6625 Everett Tower

Rank-based Estimation and Prediction for Mixed Effects Models in Nested Designs

Yusuf K. Bilgic
Western Michigan University

Hierarchical designs frequently occur in many research areas. The experimental design of interest is expressed in terms of fixed effects but for these designs, nested factors are a natural part of the experiment. These nested effects are generally considered random and must be taken into account in the statistical analysis.

Many applications have outliers in the data. Traditional analyses are sensitive to these outliers and lose power to detect the fixed effects of interest. Our goal is to investigate new rank-based methods in handling random, fixed and scale effects in k-level nested designs. We offer an algorithm, Rank Prediction Procedure (RPP) which iteratively obtains robust estimation and inference for both the fixed, scale and random effects. For simplicity, we use a 3-level nested design that deals with students nested within sections in schools. We present the results of simulation investigation, including a comparison with traditional analysis. Asymptotic derivations for our proposed estimators are our next steps.

All statistics students are expected to attend.


Department of Statistics
3304 Everett Tower
Western Michigan University
Kalamazoo MI 49008-5152 USA
(269) 387-1420 | (269) 387-1419 Fax