An official website of the United States government
Abstract Unit-level modeling strategies offer many advantages relative to the area-level models that are most often used in the context of small area estimation. For example, unit-level models aggregate naturally, allowing for estimates at any desired resolution, and also offer greater precision in many cases. We compare a variety of the methods available in the literature related to unit-level modeling for small area estimation. Specifically, to provide insight into the differences between methods, we conduct a simulation study that compares several of the general approaches. In addition, the methods used for simulation are further illustrated through an application to the American Community Survey.
more »
« less
A Comprehensive Overview of Unit-Level Modeling of Survey Data for Small Area Estimation Under Informative Sampling
Abstract Model-based small area estimation is frequently used in conjunction with survey data to establish estimates for under-sampled or unsampled geographies. These models can be specified at either the area-level, or the unit-level, but unit-level models often offer potential advantages such as more precise estimates and easy spatial aggregation. Nevertheless, relative to area-level models, literature on unit-level models is less prevalent. In modeling small areas at the unit level, challenges often arise as a consequence of the informative sampling mechanism used to collect the survey data. This article provides a comprehensive methodological review for unit-level models under informative sampling, with an emphasis on Bayesian approaches.
more »
« less
- PAR ID:
- 10447837
- Date Published:
- Journal Name:
- Journal of Survey Statistics and Methodology
- ISSN:
- 2325-0984
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Small area estimation (SAE) has become an important tool in official statistics, used to construct estimates of population quantities for domains with small sample sizes. Typical area-level models function as a type of heteroscedastic regression, where the variance for each domain is assumed to be known and plugged in following a design-based estimate. Recent work has considered hierarchical models for the variance, where the design-based estimates are used as an additional data point to model the latent true variance in each domain. These hierarchical models may incorporate covariate information but can be difficult to sample from in high-dimensional settings. Utilizing recent distribution theory, we explore a class of Bayesian hierarchical models for SAE that smooth both the design-based estimate of the mean and the variance. In addition, we develop a class of unit-level models for heteroscedastic Gaussian response data. Importantly, we incorporate both covariate information as well as spatial dependence, while retaining a conjugate model structure that allows for efficient sampling. We illustrate our methodology through an empirical simulation study as well as an application using data from the American Community Survey.more » « less
-
Abstract The Household Pulse Survey (HPS), released by the US Census Bureau at the start of the coronavirus pandemic, gathers timely information about the societal and economic impacts of coronavirus. The first phase of the survey was launched in April 2020 and ran for 12 weeks. To track the immediate impact of the pandemic, individual respondents during this phase were re-sampled for up to three consecutive weeks. Motivated by expected job loss during the pandemic, using public-use microdata, this work proposes unit-level, model-based estimators that incorporate longitudinal dependence at both the response and domain level. In particular, using a pseudo-likelihood, we consider a Bayesian hierarchical unit-level, model-based approach for both Gaussian and binary response data under informative sampling. To facilitate construction of these model-based estimates, we develop an efficient Gibbs sampler. An empirical simulation study is conducted to compare the proposed approach to models that do not account for unit-level longitudinal correlation. Finally, using public-use HPS micro-data, we provide an analysis of ‘expected job loss’ that compares both design- and model-based estimators and demonstrates superior performance for the proposed model-based approaches.more » « less
-
The Household Pulse Survey, recently released by the U.S. Census Bureau, gathers information about the respondents’ experiences regarding employment status, food security, housing, physical and mental health, access to health care, and education disruption. Design-based estimates are produced for all 50 states and the District of Columbia (DC), as well as 15 Metropolitan Statistical Areas (MSAs). Using public-use microdata, this paper explores the effectiveness of using unit-level model-based estimators that incorporate spatial dependence for the Household Pulse Survey. In particular, we consider Bayesian hierarchical model-based spatial estimates for both a binomial and a multinomial response under informative sampling. Importantly, we demonstrate that these models can be easily estimated using Hamiltonian Monte Carlo through the Stan software package. In doing so, these models can readily be implemented in a production environment. For both the binomial and multinomial responses, an empirical simulation study is conducted, which compares spatial and non-spatial models. Finally, using public-use Household Pulse Survey micro-data, we provide an analysis that compares both design-based and model-based estimators and demonstrates a reduction in standard errors for the model-based approaches.more » « less
-
Abstract Many large‐scale surveys collect both discrete and continuous variables. Small‐area estimates may be desired for means of continuous variables, proportions in each level of a categorical variable, or for domain means defined as the mean of the continuous variable for each level of the categorical variable. In this paper, we introduce a conditionally specified bivariate mixed‐effects model for small‐area estimation, and provide a necessary and sufficient condition under which the conditional distributions render a valid joint distribution. The conditional specification allows better model interpretation. We use the valid joint distribution to calculate empirical Bayes predictors and use the parametric bootstrap to estimate the mean squared error. Simulation studies demonstrate the superior performance of the bivariate mixed‐effects model relative to univariate model estimators. We apply the bivariate mixed‐effects model to construct estimates for small watersheds using data from the Conservation Effects Assessment Project, a survey developed to quantify the environmental impacts of conservation efforts. We construct predictors of mean sediment loss, the proportion of land where the soil loss tolerance is exceeded, and the average sediment loss on land where the soil loss tolerance is exceeded. In the data analysis, the bivariate mixed‐effects model leads to more scientifically interpretable estimates of domain means than those based on two independent univariate models.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1093/jssam/smad020
