A Partial Fay-Herriot model for Small Area Estimation: Estimating district-level consumption in Mozambique
Abstract
This paper proposes a new small area estimation approach that integrates Partial Least Squares within the Fay-Herriot model to address the challenges posed by high-dimensional and highly correlated auxiliary variables sets. The resulting Partial Fay-Herriot (PFH) estimator constructs supervised components that maximize their association with the target variable, enhancing model stability and predictive efficiency. Monte Carlo simulations demonstrate that PFH estimator achieves lower mean squared error than the standard Fay-Herriot estimator and outperforms principal components-based alternatives while relying on fewer latent dimensions. The methodology is applied to the estimation of district-level per capita consumption in Mozambique, where the survey data source is complemented by large set of correlated census variables. The resulting estimates highlight pronounced geographic heterogeneity and uncover spatial clusters of deprivation. Overall, the findings show that the proposed supervised dimension-reduction approach represents an effective and easily interpretable tool for producing reliable indicators in high-dimensional contexts.
Related
- Matrix Graphical Model Via Joint Estimation of Partial Correlations
- Recoverability Is a Subspace Property: A Benchmark for Certified State Estimation from Partial PDE Observations
- Characterizing Heterogeneous Rates in Finite Mixture Estimation via Partial Optimal Transport
- Optimisation of distributed storage in existing district heating networks to solve hydraulic bottlenecks
- A MINLP optimisation method for the integration of storage in district heating networks
BibTeX
@misc{spagnolo2026partial,
title = {{A Partial Fay-Herriot model for Small Area Estimation: Estimating district-level consumption in Mozambique}},
author = {Francesco Schirripa Spagnolo and Nicola Salvati and Enrico Fabrizi},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21701},
eprint = {2609.21701},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21701},
}
RIS
TY - UNPB
TI - A Partial Fay-Herriot model for Small Area Estimation: Estimating district-level consumption in Mozambique
AU - Francesco Schirripa Spagnolo
AU - Nicola Salvati
AU - Enrico Fabrizi
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21701
UR - https://arxiv.org/abs/2609.21701
AB - This paper proposes a new small area estimation approach that integrates Partial Least Squares within the Fay-Herriot model to address the challenges posed by high-dimensional and highly correlated auxiliary variables sets. The resulting Partial Fay-Herriot (PFH) estimator constructs supervised components that maximize their association with the target variable, enhancing model stability and predictive efficiency. Monte Carlo simulations demonstrate that PFH estimator achieves lower mean squared error than the standard Fay-Herriot estimator and outperforms principal components-based alternatives while relying on fewer latent dimensions. The methodology is applied to the estimation of district-level per capita consumption in Mozambique, where the survey data source is complemented by large set of correlated census variables. The resulting estimates highlight pronounced geographic heterogeneity and uncover spatial clusters of deprivation. Overall, the findings show that the proposed supervised dimension-reduction approach represents an effective and easily interpretable tool for producing reliable indicators in high-dimensional contexts.
ER -