Spatial Fay-Herriot Models for Small Area Estimation with Functional Covariates

TL;DR

This paper extends the Fay-Herriot model for small area estimation by incorporating functional covariates and spatial autocorrelation, demonstrating improved accuracy through simulation and real data analysis.

Contribution

It introduces a two-stage dimension reduction for functional covariates and integrates spatial autocorrelation modeling into the Fay-Herriot framework.

Findings

Enhanced estimation accuracy with functional covariates

Effective modeling of spatial autocorrelation improves results

Successful application to American Community Survey data

Abstract

The Fay-Herriot (FH) model is widely used in small area estimation and uses auxiliary information to reduce estimation variance at undersampled locations. We extend the type of covariate information used in the FH model to include functional covariates, such as social-media search loads or remote-sensing images (e.g., in crop-yield surveys). The inclusion of these functional covariates is facilitated through a two-stage dimension-reduction approach that includes a Karhunen-Lo\`{e}ve expansion followed by stochastic search variable selection. Additionally, the importance of modeling spatial autocorrelation has recently been recognized in the FH model; our model utilizes the intrinsic conditional autoregressive class of spatial models in addition to functional covariates. We demonstrate the effectiveness of our approach through simulation and analysis of data from the American Community Survey. We use Google Trends searches over time as functional covariates to analyze relative changes in rates of percent household Spanish-speaking in the eastern half of the United States.