Small area estimates from sample surveys / G. H. Choudhry and Y. Bélanger.: CS11-616/85-75E-PDF

"A Monte Carlo study was carried out to evaluate the performance of some alternate Small Area Estimation techniques to obtain Labour Force estimates from large scale household surveys. The Canadian Labour Force survey which utilizes a clustered multi-stage sample design has been taken as an example of a large scale household survey and the survey design was simulated using data from 1976 and 1981 censuses. The census divisions (CD's) which cut across the boundaries of design strata are taken as small areas (domains) and 1,000 Monte Carlo samples were selected to obtain LF estimates. The alternate small area estimators which include post-stratified domain, synthetic, composite and SPREE were evaluated for their Biases and Mean Square Errors. Keyfitz's (1957) variance estimator for estimating the variance of a ratio has been considered for small area estimators using population adjustment at the small area level"--Abstract.

Permanent link to this Catalogue record:
publications.gc.ca/pub?id=9.840059&sl=0

Publication information
Department/Agency Canada. Statistics Canada. Methodology Branch.
Title Small area estimates from sample surveys / G. H. Choudhry and Y. Bélanger.
Series title Working paper ; 85-75
Publication type Series - View Master Record
Language [English]
Format Electronic
Electronic document
Note(s) Digitized edition from print [produced by Statistics Canada].
"Number: CHSM 85-075E."
Includes bibliographic references.
Publishing information [Ottawa] : Statistics Canada, [1985].
Author / Contributor Choudhry, G. H.
Bélanger, Y.
Description 20 p.
Catalogue number
  • CS11-616/85-75E-PDF
Departmental catalogue number 11-616 no. 85-75
Subject terms Statistical analysis
Methodology
Surveys
Request alternate formats
To request an alternate format of a publication, complete the Government of Canada Publications email form. Use the form’s “question or comment” field to specify the requested publication.
Date modified: