E. López Vizcaíno, M. J. Lombardía Cortiña, C. Rueda Sabater
The aim of this work is to estimate the employment people in Galicia, by economic activity and quarter, using data from Labour Force Survey. Several small area model-based estimators are considered, which are derived from additive regression models. When estimators that only borrow strength across areas are considered, the trend pattern across time depends on the characteristics of the area and the model used to derive the estimators. Often, for a given area and model, the time trend pattern is more abrupt than desired. On the other side, when time series models are considered too smoothed patterns are generated and prevent major trend changes being detected. We propose a very different and simple strategy to borrow strength across time, as well as across areas, that lacks these drawbacks. The proposal makes use of a Generalized Akaike Information Criteria (GAIC) measure. The idea is select the model using the area-specific component of an aggregated GAIC measure across time.
Palabras clave: AIC, small area, Galicia, Fay-Herriot, spline model, monotone model
Programado
M05.4 Sesión de la Sociedad Gallega para la Promoción de la Estadística y la Investigación Operativa
6 de septiembre de 2016 11:00
Aula 21.07
