Small-area estimates using model-based methods

In small areas, the number of sample plots is usually small, and the classical estimators have a large variance. Information from nearby areas can be utilized to improve the subarea estimates using either nonparametric or parametric models. In this study, a number of model-based estimators for small-area estimation are presented. To illustrate the presented methods a numerical example in a real inventory situation is given. The auxiliary information used in this study is pure coordinate information, but the methods are applicable also for other kinds of auxiliary information. The object of this study is to compare the features of the presented small-area estimation methods and to discuss the applicability of these methods in different situations.

Download Full-text

A comparison of small-area estimation techniques to estimate selected stand attributes using LiDAR-derived auxiliary variables

Canadian Journal of Forest Research ◽

10.1139/x11-033 ◽

2011 ◽

Vol 41 (6) ◽

pp. 1189-1201 ◽

Cited By ~ 25

Author(s):

Michael E. Goerndt ◽

Vicente J. Monleon ◽

Hailemariam Temesgen

Keyword(s):

Small Area ◽

Small Area Estimation ◽

Auxiliary Information ◽

Estimation Methods ◽

Stem Volume ◽

Small Areas ◽

Forest Density ◽

Area Estimation ◽

Forest Attributes ◽

Areas Of Interest

One of the challenges often faced in forestry is the estimation of forest attributes for smaller areas of interest within a larger population. Small-area estimation (SAE) is a set of techniques well suited to estimation of forest attributes for small areas in which the existing sample size is small and auxiliary information is available. Selected SAE methods were compared for estimating a variety of forest attributes for small areas using ground data and light detection and ranging (LiDAR) derived auxiliary information. The small areas of interest consisted of delineated stands within a larger forested population. Four different estimation methods were compared for predicting forest density (number of trees/ha), quadratic mean diameter (cm), basal area (m2/ha), top height (m), and cubic stem volume (m3/ha). The precision and bias of the estimation methods (synthetic prediction (SP), multiple linear regression based composite prediction (CP), empirical best linear unbiased prediction (EBLUP) via Fay–Herriot models, and most similar neighbor (MSN) imputation) are documented. For the indirect estimators, MSN was superior to SP in terms of both precision and bias for all attributes. For the composite estimators, EBLUP was generally superior to direct estimation (DE) and CP, with the exception of forest density.

Download Full-text

Applications of Small Area Estimation to Generalization With Subclassification by Propensity Scores

Journal of Educational and Behavioral Statistics ◽

10.3102/1076998617733828 ◽

2017 ◽

Vol 43 (2) ◽

pp. 182-224

Author(s):

Wendy Chan

Keyword(s):

Propensity Score ◽

Small Area ◽

Propensity Scores ◽

Small Area Estimation ◽

Auxiliary Information ◽

Population Data ◽

Small Sample ◽

Estimation Methods ◽

Sample Sizes ◽

Area Estimation

Policymakers have grown increasingly interested in how experimental results may generalize to a larger population. However, recently developed propensity score–based methods are limited by small sample sizes, where the experimental study is generalized to a population that is at least 20 times larger. This is particularly problematic for methods such as subclassification by propensity score, where limited sample sizes lead to sparse strata. This article explores the potential of small area estimation methods to improve the precision of estimators in sparse strata using population data as a source of auxiliary information to borrow strength. Results from simulation studies identify the conditions under which small area estimators outperform conventional estimators and the limitations of this application to causal generalization studies.

Download Full-text

Small Area Estimation of Smoke-free Workplace Polices and Home Rules in U.S. Counties

Nicotine & Tobacco Research ◽

10.1093/ntr/ntab015 ◽

2021 ◽

Author(s):

Benmei Liu ◽

Isaac Dompreh ◽

Anne M Hartman

Keyword(s):

Tobacco Control ◽

Small Area ◽

Small Sample Size ◽

Health Hazard ◽

Small Area Estimation ◽

Small Sample ◽

County Level ◽

Area Estimation ◽

Model Based ◽

Workplace Policies

Abstract Background The workplace and home are sources of exposure to secondhand smoke (SHS), a serious health hazard for nonsmoking adults and children. Smoke-free workplace policies and home rules protect nonsmoking individuals from SHS and help individuals who smoke to quit smoking. However, estimated population coverages of smoke-free workplace policies and home rules are not typically available at small geographic levels such as counties. Model-based small area estimation techniques are needed to produce such estimates. Methods Self-reported smoke-free workplace policies and home rules data came from the 2014-2015 Tobacco Use Supplement to the Current Population Survey. County-level design-based estimates of the two measures were computed and linked to county-level relevant covariates obtained from external sources. Hierarchical Bayesian models were then built and implemented through Markov Chain Monte Carlo methods. Results Model-based estimates of smoke-free workplace policies and home rules were produced for 3,134 (out of 3,143) U.S. counties. In 2014-2015, nearly 80% of U.S. adult workers were covered by smoke-free workplace policies, and more than 85% of U.S. adults were covered by smoke-free home rules. We found large variations within and between states in the coverage of smoke-free workplace policies and home rules. Conclusions The small-area modeling approach efficiently reduced the variability that was attributable to small sample size in the direct estimates for counties with data and predicted estimates for counties without data by borrowing strength from covariates and other counties with similar profiles. The county-level modeled estimates can serve as a useful resource for tobacco control research and intervention. Implications Detailed county- and state-level estimates of smoke-free workplace policies and home rules can help identify coverage disparities and differential impact of smoke-free legislation and related social norms. Moreover, this estimation framework can be useful for modeling different tobacco control variables and applied elsewhere, e.g., to other behavioral, policy, or health related topics.

Download Full-text

District level child nutrition status in Bangladesh: An application of small area estimation method

10.47302/jsr.2019530103 ◽

2019 ◽

Vol 53 (1) ◽

pp. 45-61

Author(s):

Mossamet Kamrun Nesa

Keyword(s):

Small Area ◽

Child Nutrition ◽

Small Area Estimation ◽

Estimation Method ◽

Nutrition Status ◽

Estimation Methods ◽

Small Scale ◽

Demographic Health Survey ◽

Area Estimation ◽

National Surveys

National level indicators of child undernutrition often hide the real scenario across a country. In order to construct a child nutrition map, accurate estimates of undernutrition are required at very small spatial scales, typically the administrative units of a country or a region within a country. Although comprehensive data on child nutrition are collected in national surveys, the small scale estimates cannot be calculated using the standard estimation methods employed in national surveys, since such methods are designed to produce national or regional level estimates, and assume large samples. Small area estimation method has been widely used to find such micro-level estimates. Due to lack of unit level data, area level small area estimation methods (e.g., Fay-Herriot method) are widely used to calculate small-scale estimates. In Bangladesh, a few works have been done to estimate district level child nutrition status. The Bangladesh Demographic Health Survey covers all districts but district wise sample sizes are very small to get consistent estimates. In this paper, Fay-Herriot Model has been developed to calculate district wise estimates with efficient mean squared error. The Bangladesh Demographic Health Survey 2011 and Population Census 2011 are utilized for this study.

Download Full-text

Simple House Needs in Jember with Robust Small Area Estimation

Jurnal ILMU DASAR ◽

10.19184/jid.v18i1.3159 ◽

2017 ◽

Vol 18 (1) ◽

pp. 1

Author(s):

Frida Murtinasari ◽

Alfian Futuhul Hadi ◽

Dian Anggraeni

Keyword(s):

Small Area ◽

Robust Regression ◽

Small Sample Size ◽

Small Area Estimation ◽

Small Sample ◽

Estimation Methods ◽

Linear Unbiased Prediction ◽

Area Estimation ◽

Best Linear Unbiased ◽

M Estimation

SAE (Small Area Estimation) is often used by researchers, especially statisticians to estimate parameters of a subpopulation which has a small sample size. Empirical Best Linear Unbiased Prediction (EBLUP) is one of the indirect estimation methods in Small Area Estimation. The presence of outliers in the data can not guarantee that these methods yield precise predictions . Robust regression is one approach that is used in the model Small Area Estimation. Robust approach in estimating such a small area known as the Robust Small Area Estimation. Robust Small Area Estimation divided into several approaches. It calls Maximum Likelihood and M- Estimation. From the result, Robust Small Area Estimation with M-Estimation has the smallest RMSE than others. The value is 1473.7 (with outliers) and 1279.6 (without outlier). In addition the research also indicated that REBLUP with M-Estimation more robust to outliers. It causes the RMSE value with EBLUP has five times to be large with only one outlier are included in the data analysis. As for the REBLUP method is relatively more stable RMSE results.

Download Full-text

MEMODELKAN TINGKAT PENGANGGURAN DI KOTA DENPASAR DENGAN PENDUGAAN AREA KECIL EMPIRICAL BAYES

E-Jurnal Matematika ◽

10.24843/mtk.2021.v10.i02.p325 ◽

2021 ◽

Vol 10 (2) ◽

pp. 81

Author(s):

REYNALDO PANJI WICAKSONO ◽

I KOMANG GDE SUKARSA ◽

I PUTU EKA NILA KENCANA

Keyword(s):

Economic Development ◽

Unemployment Rate ◽

Small Area ◽

Empirical Bayes ◽

Small Area Estimation ◽

Economic Conditions ◽

Small Areas ◽

Area Estimation ◽

Direct Estimation ◽

Selection Of

Economic development are described by the unemployment rate. The higher unemployment rate, the weaker economic conditions. Nowadays more policies require information on small areas. The direct estimation does not provide accurate results in smaller areas. Thus the small area estimation becomes an alternative to estimate the parameters. The accuracy depends on the selection of the predictors. In 2019, the unemployment rate in Denpasar is 2,22%. The result shows that the unemployment rate in each district in Denpasar varies from 0,1% to 0,3%

Download Full-text

Adult diet in England: where is more support needed to achieve dietary recommendations?

10.21203/rs.3.rs-67773/v1 ◽

2020 ◽

Author(s):

Dianna Smith ◽

Christina Vogel ◽

Monique Campbell ◽

Nisreen Alwan ◽

Graham Moon

Keyword(s):

Urban Areas ◽

Small Area ◽

Vegetable Consumption ◽

Small Area Estimation ◽

Estimation Methods ◽

Dietary Recommendations ◽

Nutrition Survey ◽

Targeted Interventions ◽

Area Estimation ◽

Adult Diet

Abstract Background: Small-area estimation models are regularly commissioned by public health bodies to identify areas of greater inequality and target areas for intervention in a range of behaviours and outcomes. Such local modelling has not been completed for diet consumption in England despite diet being an important predictor of health status. The study sets out whether aspects of adult diet can be modelled from previously collected data to define and evaluate area-level interventions to address obesity and ill-health.Methods: Adults aged 16 years and over living in England. Consumption of fruit, vegetables, and sugar-sweetened beverages (SSB) are modelled using small-area estimation methods in English neighbourhoods (Middle Super Output Areas [MSOA]) to identify areas where reported portions are significantly different from recommended levels of consumption. The selected aspects of diet are modelled from respondents in the National Diet and Nutrition Survey using pooled data from 2008-2016.Results: Estimates indicate that the average prevalence of adults consuming less than one portion of fruit, vegetables or 100% juice each day by MSOA is 6.9% (range of 4.3 to 14.7%, SE 0.06) and the average prevalence of drinking more than 330ml/day of SSB is 11.5% (range of 5.7 to 30.5%, SE 0.03). Credible intervals around the estimates are wider for SSB consumption. The results identify areas including regions in London and urban areas in the North of England which may be prioritised for targeted interventions to support reduced consumption of SSB and/or an increase in portions of fruit and vegetables.Conclusion: These estimates provide valuable information at a finer spatial scale than is presently feasible, allowing for within-country and locality prioritisation of resources to improve diet. Local, targeted interventions to improve fruit and vegetable consumption such as subsidies or voucher schemes should be considered where consumption of these foods is predicted to be low.

Download Full-text

A COMPARISON OF SMALL AREA ESTIMATION METHODS FOR POVERTY MAPPING

Statistics in Transition New Series ◽

10.21307/stattrans-2016-005 ◽

2016 ◽

Vol 17 (1) ◽

pp. 41-66 ◽

Cited By ~ 6

Author(s):

María Guadarrama ◽

Isabel Molina ◽

J. N. K. Rao

Keyword(s):

Small Area ◽

Small Area Estimation ◽

Estimation Methods ◽

Area Estimation ◽

Poverty Mapping

Download Full-text

A Study of Small Area Estimation for Italian Structural Business Statistics

Journal of Official Statistics ◽

10.2478/jos-2018-0025 ◽

2018 ◽

Vol 34 (2) ◽

pp. 543-555

Author(s):

Orietta Luzi ◽

Fabrizio Solari ◽

Fabiana Rocci

Keyword(s):

Small Area ◽

Small Area Estimation ◽

Auxiliary Information ◽

Optimal Estimation ◽

Estimation Strategy ◽

Area Estimation ◽

Statistical Institute ◽

Business Statistics ◽

Different Types ◽

Quality And Reliability

Abstract The Frame SBS is a statistical register which has been developed at the Italian National Statistical Institute to support the annual estimation of structural business statistics (SBS). Actually, a number of core SBS are estimated by combining microdata directly supplied by different administrative sources. In this context, more accurate estimates for those SBS that are not covered by administrative sources can be obtained through small area estimation (SAE). In this article, we illustrate an application of SAE methods in the framework of the Frame SBS register in order to assess the potential advantages that can be achieved in terms of increased quality and reliability of the target variables. Different types of auxiliary information and approaches are compared in order to identify the optimal estimation strategy in terms of precision of the estimates.

Download Full-text