Skip to main content
SLU publication database (SLUpub)

Report2015

Small area estimation under a multivariate linear model for repeated measures data.

Ngaruye, Innocent; Nzabanita, Joseph; Von Rosen, Dietrich; von Rosen, Dietrich; Singull, Martin

Abstract

In this paper, we consider small area estimation under a multivariatelinear regression model for repeated measures data.  The aim of theproposed model is to get a model which borrows strength across smallareas and over time, by incorporating simultaneously the area effectsand time correlation.  The model accounts for repeated surveys, groupindividuals  and  random  effects  variations.   Estimation  of  model  pa-rameters  is  discussed  within  a  restricted  maximum  likelihood  basedapproach.   Prediction  of  random  effects  and  the  prediction  of  smallarea means across time points and per group units for all time pointsare derived.  The results are supported by a simulation study.

Published in

LiTH-MAT-R
2015, number: 2015/09Publisher: Linköping University