scholarly journals Small area statistics and quality management –the Polish perspective

2018 ◽  
Vol 16 (22) ◽  
pp. 37-54
Author(s):  
Jan Kordos
Keyword(s):  
Quality Management ◽  
Small Area ◽  
Small Area Statistics

The Estimation of Population Microdata by Using Data from Small Area Statistics and Samples of Anonymised Records

1998 ◽  
Vol 30 (5) ◽  
pp. 785-816 ◽  
Author(s):  
P Williamson ◽  
M Birkin ◽  
P H Rees
Keyword(s):  
Small Area ◽  
Census Data ◽  
Population Data ◽  
Combinatorial Optimisation ◽  
Population Characteristics ◽  
Test Statistics ◽  
Enumeration District ◽  
Small Area Statistics ◽  
Using Data ◽  
Combining Information

Census data can be represented both as lists and as tabulations of household/individual attributes. List representation of Census data offers greater flexibility, as the exploration of interrelationships between population characteristics is limited only by the quality and scope of the data collected. Unfortunately, the released lists of household/individual attributes (Samples of Anonymised Records, SARs) are spatially referenced only to areas (single or merged districts) with populations of 120 000 or more, whereas released tabulations are available for units as small as single enumeration districts (Small Area Statistics, SAS). Intuitively, it should be possible to derive list-based estimates of enumeration district populations by combining information contained in the SAR and the SAS. In this paper we explore the range of solutions that could be adapted to this problem which, ultimately, is presented as a complex combinatorial optimisation problem. Various techniques of combinatorial optimisation are tested, and preliminary results from the best performing algorithm are evaluated. Through this process, the lack of suitable test statistics for the comparison of observed and expected tabulations of population data is highlighted.

PEMODELAN EMPIRICAL BEST LINEAR UNBIASED PREDICTION (EBLUP) DALAM PENDUGAAN AREA KECIL

2020 ◽  
Vol 2019 (1) ◽  
pp. 117-123
Author(s):  
M Irsyad Ilham
Keyword(s):  
Small Area ◽  
Best Linear Unbiased Prediction ◽  
Best Linear Unbiased Predictor ◽  
Linear Unbiased Prediction ◽  
Indirect Estimation ◽  
Direct Estimation ◽  
Small Area Statistics ◽  
Best Linear Unbiased ◽  
Unbiased Prediction

Kebutuhan terhadap data pada level mikro semakin tinggi. Di sisi lain, Badan Pusat Statistik (BPS) membutuhkan biaya yang cukup besar untuk pengumpulan data seiring dengan banyaknya kegiatan survei rutin yang dilakukan berulang kali. Hal ini menunjukkan BPS memerlukan metode estimasi yang akan menghasilkan statistik yang efektif dan efisien, di samping hemat dan menghasilkan statistik dengan ketelitian yang memadai. Penelitian ini akan menerapkan metode Small Area Statistics (SAE) atau estimasi wilayah kecil untuk keperluan estimasi rata-rata pengeluran per kapita menurut kelurahan/desa di kabupaten Sukamara. Metode estimasi level kelurahan/desa menggunakan model Empirical Best Linear Unbiased Predictor (EBLUP). Tahapan pertama yakni melakukan pendugaan langsung (direct estimation) nilai rata-rata pengeluaran per kapita pada desa yang terpilih menjadi sampel. Selanjutnya, dilakukan estimasi tidak langsung (indirect estimation) untuk mengestimasi pengeluaran per kapita seluruh desa di kabupaten Sukamara. Metode tidak langsung didasarkan pada pemodelan regresi linier dengan menggunakan variabel tambahan (auxiliary variabel) yang memiliki hubungan yang kuat dan linier terhadap variabel prediktor. Variabel tambahan yang digunakan dalam penelitian ini adalah jumlah keluarga non listrik, jumlah sarana pendidikan, jumlah sarana kesehatan, jumlah kasus gizi buruk, jumlah keluarga dengan Surat Keterangan Tidak Mampu (SKTM), dan jumlah keluarga yang memiliki jamkesmas/askes. Dari hasil pemodelan, secara umum pengeluaran per kapita per bulan dari setiap desa berada di atas satu juta rupiah. Penduduk desa-desa yang terletak di kecamatan Balai Riam dan Permata Kecubung terlihat banyak mengeluarkan sejumlah dana, untuk keperluan makanan maupun non-makanan. Hal ini diindikasikan karena desa-desa tersebut yang memiliki infrastruktur yang memadai ke Kota Pangkalan Bun.

Estimation the number of economically inactive people in poviats of the Kujawsko-Pomorskie Voivodship using methods of small area statistics

2016 ◽  
Vol 61 (5) ◽  
pp. 37-47
Author(s):  
Marek Ręklewski ◽  
Dominik Śliwicki
Keyword(s):  
Small Area ◽  
Estimation Methods ◽  
Low Level ◽  
Small Area Statistics ◽  
Level Of Aggregation

The aim of this paper is to estimate the number of economically inactive people at a level of poviats in Kujawsko-Pomorskie Voivodship using selected methods of small area statistics and to assess the precision of the estimates obtained. The analysis is an example of the use of selected small area estimators to estimate the number of economically inactive people in poviats of Kujawsko-Pomorskie Voivodship. The obtained results allow to inference about the size of the economically inactive population at a relatively low level of aggregation. Previously used in public statistics estimation methods did not give the possibility to obtain data at this level due to the large errors of estimates.

Bringing the Missing Million Home: Correcting the 1991 Small Area Statistics for Undercount

2002 ◽  
Vol 34 (6) ◽  
pp. 1021-1035 ◽  
Author(s):  
Richard Mitchell ◽  
Danny Dorling ◽  
David Martin ◽  
Ludi Simpson
Keyword(s):  
Small Area ◽  
Census Data ◽  
Decennial Census ◽  
Data Set ◽  
Web Based ◽  
Marginalised Groups ◽  
Small Area Statistics ◽  
Web Based System

The 1991 UK Decennial Census missed about 1.2 million people. These missing individuals present a serious challenge to any census user interested in measuring intercensal change, particularly amongst the most marginalised groups in society who were prominent amongst the missing population. Recently, a web-based system for accessing census data from the 1971, 1981, and 1991 censuses was launched ( www.census.ac.uk/cdu/lct ). The ‘LCT’ package also provides access to a set of 1991 small area statistics (SAS) which have been corrected to compensate for the missing million. The authors explain the methods used for adjusting the SAS counts, provide examples of the differences between analysis with the adjusted and unadjusted data, and recommend the use of the new data set to all those interested in intercensal change.

Sign in / Sign up

Export Citation Format

Share Document