EFFECTS OF SAMPLING VARIANCE ESTIMATION METHODS ON PRECISION OF SMALL AREA ESTIMATION

2017 ◽  
Vol 14 (2) ◽  
pp. 93-106
Author(s):  
Mojtaba Soltani-Kermanshahi ◽  
Yadollah Mehrabi ◽  
Amir Kavousi ◽  
Ahmad-Reza Baghestani ◽  
Fatemeh Mohammadi-Nasrabadi
Keyword(s):  
Small Area ◽  
Variance Estimation ◽  
Small Area Estimation ◽  
Estimation Methods ◽  
Sampling Variance ◽  
Area Estimation

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

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.

scholarly journals Simple House Needs in Jember with Robust Small Area Estimation

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.

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

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.

scholarly journals A COMPARISON OF SMALL AREA ESTIMATION METHODS FOR POVERTY MAPPING

2016 ◽  
Vol 17 (1) ◽  
pp. 41-66 ◽  
Author(s):  
María Guadarrama ◽  
Isabel Molina ◽  
J. N. K. Rao
Keyword(s):  
Small Area ◽  
Small Area Estimation ◽  
Estimation Methods ◽  
Area Estimation ◽  
Poverty Mapping

scholarly journals A Map of the Poor or a Poor Map?

Mathematics ◽  
2021 ◽  
Vol 9 (21) ◽  
pp. 2780
Author(s):  
Paul Corral ◽  
Kristen Himelein ◽  
Kevin McGee ◽  
Isabel Molina
Keyword(s):  
Small Area ◽  
Census Data ◽  
Mean Squared Error ◽  
Small Area Estimation ◽  
Selection Procedure ◽  
Data Transformation ◽  
Estimation Methods ◽  
Simulation Experiments ◽  
Area Estimation ◽  
Level Information

This paper evaluates the performance of different small area estimation methods using model and design-based simulation experiments. Design-based simulation experiments are carried out using the Mexican Intra Censal survey as a census of roughly 3.9 million households from which 500 samples are drawn using a two-stage selection procedure similar to that of Living Standards Measurement Study (LSMS) surveys. The estimation methods considered are that of Elbers, Lanjouw and Lanjouw (2003), the empirical best predictor of Molina and Rao (2010), the twofold nested error extension presented by Marhuenda et al. (2017), and finally an adaptation, presented by Nguyen (2012), that combines unit and area level information, and which has been proposed as an alternative when the available census data is outdated. The findings show the importance of selecting a proper model and data transformation so that model assumptions hold. A proper data transformation can lead to a considerable improvement in mean squared error (MSE). Results from design-based validation show that all small area estimation methods represent an improvement, in terms of MSE, over direct estimates. However, methods that model unit level welfare using only area level information suffer from considerable bias. Because the magnitude and direction of the bias is unknown ex ante, methods relying only on aggregated covariates should be used with caution, but may be an alternative to traditional area level models when these are not applicable.

scholarly journals Estimating poverty and unemployment using small area estimation methods

2021 ◽  
Author(s):  
Ελένη Μαλαπάνη
Keyword(s):  
Small Area ◽  
Small Area Estimation ◽  
Estimation Methods ◽  
Area Estimation ◽  
Area Model

Τις τελευταίες δεκαετίες η καταπολέμηση της φτώχειας και της ανεργίας αποτελεί μία από τις κυριότερες προκλήσεις της Ευρώπης και όχι μόνο. Η Ευρωπαϊκή Επιτροπή στα πλαίσια της στρατηγικής «Ευρώπη 2020» έθεσε το παραπάνω πρόβλημα ως έναν από τους βασικούς στόχους της. Για την εκτίμηση χαρακτηριστικών όπως η φτώχεια και η ανεργία τα κράτη έχουν σχεδιάσει ειδικές έρευνες οι οποίες δίνουν μεν αξιόπιστες πληροφορίες σε εθνικό επίπεδο, αλλά λόγω σχεδιασμού και μεγέθους των δειγμάτων δεν μπορούν να δώσουν αντίστοιχες αξιόπιστες εκτιμήσεις σε μικρότερες γεωγραφικές περιοχές όπως π.χ. οι νομοί (NUTS 3) και οι δήμοι. Για τη διαχείριση και επίλυση του παραπάνω προβλήματος, δηλαδή την επίτευξη στατιστικών εκτιμήσεων κοινωνικών και οικονομικών δεικτών σε μικρές γεωγραφικές περιοχές, προτείνεται ο συνδυασμός των παραπάνω τύπων δεδομένων (ετήσιες έρευνες και απογραφικά δεδομένα). Ο συνδυασμός αυτός μπορεί να επιτευχθεί με τη χρήση προηγμένων στατιστικών μεθόδων που παράγουν εκτιμήσεις σε μικρές γεωγραφικές περιοχές και έχουν τη γενική ονομασία «Small Area Estimation» (SAE). Στην Ελλάδα το μικρότερο γεωγραφικό επίπεδο για το οποίο δίνονται εκτιμήσεις της φτώχειας και της ανεργίας είναι αυτό των περιφερειών. Ο κύριος στόχος αυτής της διατριβής είναι να αναπτύξει και να παράσχει αξιόπιστες εκτιμήσεις για τη φτώχεια και την ανεργία στην Ελλάδα σε μικρότερο γεωγραφικό επίπεδο από αυτό των περιφερειών, δηλαδή σε επίπεδο Νομών (NUTS 3) χρησιμοποιώντας τις μεθόδους SAE. Τα υπό εκτίμηση χαρακτηριστικά είναι το ποσοστό της φτώχειας, το χάσμα της φτώχειας καθώς και το ποσοστό της ανεργίας του ελληνικού πληθυσμού σε δύο διαφορετικές χρονικές στιγμές, το 2009 (λίγο πριν από την έναρξη της ελληνικής χρηματοπιστωτικής κρίσης) και το 2013 (κατά τη διάρκεια της κρίσης)). Για την επίτευξη των παραπάνω υιοθετήθηκε ο εκτιμητής EBLUP με βάση το μοντέλο Fay and Herriot, συνδυάζοντας δεδομένα από την έρευνα EU-SILC 2009 και 2013 με βοηθητικά δεδομένα από την εθνική απογραφή του 2001 και 2011, αντίστοιχα. Συγκεκριμένα εξετάσθηκαν και αναλύθηκαν 19 βοηθητικές μεταβλητές από την απογραφή του 2001 και 32 βοηθητικές μεταβλητές από την απογραφή του 2011. Προκειμένου να κατασκευαστεί το βέλτιστο μοντέλο μικρής περιοχής (small area model) για κάθε ένα από τα υπό εκτίμηση χαρακτηριστικά, χρησιμοποιήθηκε μια διαδικασία τριών φάσεων για την επιλογή των τελικών βοηθητικών μεταβλητών. Έπειτα διάφοροι διαγνωστικοί έλεγχοι εφαρμόστηκαν με σκοπό την αξιολόγηση της καταλληλόλητας και απόδοσης των επιλεγμένων SAE μοντέλων καθώς και της αξιοπιστίας των αποτελεσμάτων. Τα αποτελέσματα αυτών των διαγνωστικών ελέγχων έδειξαν ότι τα επιλεγμένα μοντέλα παρέχουν καλή προσαρμογή στα δεδομένα καθώς και αξιόπιστες εκτιμήσεις. Τα αποτελέσματα της έρευνας ήταν ιδιαίτερα ενθαρρυντικά καθώς η εφαρμογή των μεθόδων SAE πέτυχε ένα στατιστικά σημαντικό συνολικό κέρδος απόδοσης τόσο για την εκτίμηση της φτώχειας όσο και της ανεργίας έναντι των άμεσων εκτιμητών. Συγκεκριμένα, τα αποτελέσματα έδειξαν μία στατιστικά σημαντική μείωση τόσο των τιμών του συντελεστή μεταβλητότητας (CV) όσο και των τιμών του μέσου τετραγωνικού σφάλματος (MSE) του EBLUP εκτιμητή με βάση το μοντέλο F-H έναντι των άμεσων εκτιμητών σχεδόν σε όλους τους νομούς. Η μείωση ήταν αισθητά μεγαλύτερη στους νομούς με μικρό μέγεθος δείγματος. Επίσης, τα αποτελέσματα των εκτιμήσεων τόσο της φτώχειας όσο και της ανεργίας έδειξαν σημαντικές διαφορές στο χάρτη της Ελλάδας τις χρονιές 2009 και 2013. Η παρούσα μελέτη συμβάλλει στην ολοένα και αυξανόμενη ζήτηση για εκτιμήσεις κοινωνικών χαρακτηριστικών σε μικρές γεωγραφικές περιοχές αναπτύσσοντας κατάλληλα SAE μοντέλα και δίνοντας εκτιμήσεις για τη φτώχεια και την ανεργία στην Ελλάδα για πρώτη φορά σε επίπεδο νομών. Οι εκτιμήσεις αυτές μπορούν να συμβάλουν στη διαμόρφωση και στόχευση πολιτικών για τη σωστή κατανομή των δημόσιων κονδυλίων σε μικρές γεωγραφικές περιοχές.

A Study of Disability Across Districts of the Province of Wielkopolska Using Small Area Estimation Methods

2019 ◽  
Vol 65 (4) ◽  
pp. 449-472
Author(s):  
Tomasz Klimanek ◽  
Marcin Szymkowiak ◽  
Marcin Szymkowiak ◽  
Tomasz Józefowski
Keyword(s):  
Small Area ◽  
Small Area Estimation ◽  
Estimation Methods ◽  
Sources Of Information ◽  
Estimation Errors ◽  
Area Estimation ◽  
Central Statistical ◽  
Central Statistical Office ◽  
Level Of Aggregation ◽  
High Level

Surveys and censuses conducted by the Central Statistical Office in Poland are the main sources of information about disability for official statistics. Because sample sizes for relevant cross-classification domains are too small to employ classical estimation methods, results are usually published at a relatively high level of aggregation (at country or province level) or for very broadly defined domains. To meet the growing demand for detailed information about disability, the authors present an attempt of applying the methodology of small area estimation to estimate the percentage of disabled people, in the legal and biological sense, across districts (NUTS 4/LAU 1 units) of the province of Wielkopolska crossclassified by the level of education. This methodological exercise is based on data from the 2011 census and employs selected techniques of indirect estimation. Estimates obtained in the study provide an indication of the spatial variation of disability in the target domains with greater precision. It is worth noting that this level of aggregation has not been considered for purposes of official statistical outputs because of unacceptably high estimation errors of the direct estimator.

scholarly journals Adult diet in England: Where is more support needed to achieve dietary recommendations?

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0252877
Author(s):  
Dianna M. 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

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, urban areas in the North of England and the South coast 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.

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

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.

Sign in / Sign up

Export Citation Format

Share Document