Tatjana von Rosen: Spatio-Temporal Mixed Linear Models in Small Area Estimation

Small area estimation has received a lot of attention due to its applications in official statistics. Mixed linear models have been used in small area estimation due to their flexibility in modeling between and within unit correlations and handling di fferent sources of variation. Longitudinal data are usually collected in order to get information about changes over time. We consider small area estimation from longitudinal survey data, where the sampled units are observed at several time points. Area-level mixed linear models that take into account possible correlation among the neighboring areas and time points (spatio-temporal mixed linear models) are studied. The covariance structures suitable for describing spatio-temporal dependence are also discussed.