Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/31123 
Year of Publication: 
2005
Series/Report no.: 
Discussion Paper No. 442
Publisher: 
Ludwig-Maximilians-Universität München, Sonderforschungsbereich 386 - Statistische Analyse diskreter Strukturen, München
Abstract: 
When some observations in the sample data are missing, the application of the regression method is considered for the estimation of population mean with and without the use of imputation. The performance properties of the estimators based on the methods of mean imputation, regression imputation and no imputation are analyzed and the superiority of one method over the other is examined.
Subjects: 
missing data mechanism
regression analysis
generalized additive models
imputation
MSE-superiority
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size
190.01 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.