Premium
Estimation Methods and Related Systems at Statistics Canada
Author(s) -
Singh M.P,
Hidiroglou M.A.,
Gambino J.G.,
Kovaçević M.S.
Publication year - 2001
Publication title -
international statistical review
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.051
H-Index - 54
eISSN - 1751-5823
pISSN - 0306-7734
DOI - 10.1111/j.1751-5823.2001.tb00470.x
Subject(s) - small area estimation , estimation , estimator , imputation (statistics) , weighting , econometrics , statistics , data quality , computer science , geography , mathematics , missing data , engineering , medicine , metric (unit) , operations management , systems engineering , radiology
Summary This paper provides an overview of research in estimation techniques, their application, and the development of generalized estimation systems at Statistics Canada. In Canada, the demand for more detailed and better quality cross‐sectional data related to various sodo‐economic issues has increased significantly in recent years. Also, there has been increasing interest in longitudinal data to better understand and interpret the relationships among variables, necessitating the implementation of a number of large scale panel surveys by Statistics Canada. The paper briefly discusses estimation for longitudinal data and a weighting approach developed for cross‐sectional data Prom these surveys. For cross‐sectional household and business surveys, as well as the census of population, appropriate dibration estimators developed for each situation are briefly discussed. In addition, regression composite estimation, a method developed to improve the quality of cross‐sectional estimates from rotating panel surveys such as the Canadian Labour Force Survey, is presented. With regard to more detailed cross‐sectional estimates at sub‐provincial levels, different approaches to small area estimation developed for various programs are also presented. We SUmmarize the various modules developed lor the GeneraIiized Ektimation System. important new developments within the system include two‐phase estimation as well as the estimation of variance for a number of imputation procedures. We briefly review the status of current estimation research on selected topics as well as the direction of future research.