B0454
Title: Small area estimation of economic indicators under unit-level generalized additive models for location, scale and shape
Authors: lorenzo mori - University of Bologna (Italy) [presenting]
Maria Rosaria Ferrante - University of Bologna (Italy)
Abstract: A small area estimation (SAE) unit-level model is proposed based on generalized additive models for location, scale and shape (GAMLSS). GAMLSS at first completely release the exponential family distribution assumption for the response variable replacing it with a distribution, that includes highly skewed and/or kurtotic continuous and discrete distributions. Secondly, GAMLSS give the opportunity to model each distributional parameter depending on covariates leading to borrowing strength not only by location parameter but potentially also form covariates explaining other model parameters (scale and/or shape). A parametric bootstrap approach to estimate MSE is proposed. The performance of the proposed estimators is evaluated based on Monte Carlo simulations in both design-based and model-based frameworks. The results obtained show that the proposed small-area predictors work well with respect to the well-known EBLUP unit-level SAE estimator. Based on SAE-GAMLSS per-capita consumption of Italian and foreign households in Italian regions, in urban and rural areas, is estimated. Results show that the well-known Italian North-South divide does not hold for foreigners.
