Comparison of benchmarking methods with and without survey error modelling: CS11-617/2003-6E-PDF
"In this paper, we will show the relationship among three types of benchmarking methods in the literature, namely the Denton (original and modified), the regression, and the signal extraction methods. Assuming the survey error series follows an AR(1) model, by simulation, we investigate the impacts of mis-specification of the model on the benchmarking predictions based on the criterion of minimizing root-mean-squared error of prediction. It is concluded that the survey error modelling procedure proposed by Chen and Wu (2000, 2001) may lead to results as good as those obtained from using the true survey error models"--Abstract.
|Department/Agency||Statistics Canada. Methodology Branch.|
|Title||Comparison of benchmarking methods with and without survey error modelling|
|Series Title||Working paper ;|
|Publication Type||Series - View Master Record|
|Electronic Document|| |
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact the authoring department to request a format other than those available.
We invite you to consult the Frequently Asked Questions page for additional information regarding the Archived Content notice.
Note: The URLs contained in this/these document(s) may no longer be functional
|Note||Digitized edition from print [produced by Statistics Canada]. "BSMD-2003-006E." "November 2003."|
|Number of Pages||23 p.|
|Departmental Catalogue Number||11-617-XPE no. 0306|
|Subject Terms||Statistical analysis, Methodology, Surveys|
- Date modified: