Comparison of predictive mean matching and regression multiple imputation methods using the 2006 Family Income and Expenditure Survey (FIES)
Date of Publication
2010
Document Type
Bachelor's Thesis
Degree Name
Bachelor of Science in Statistics Major in Actuarial Science
Subject Categories
Physical Sciences and Mathematics
College
College of Science
Department/Unit
Mathematics and Statistics
Thesis Adviser
Rechel G. Arcilla
Defense Panel Chair
Arturo Y. Pacificador, Jr.
Defense Panel Member
Shirlee R. Ocampo
Andrew Philip A. Wee
Abstract/Summary
This study used the 2006 Family Income and Expenditure Survey (FIES) data of the National Capital Region (NCR) in comparing the two multiple imputation methods predictive mean matching method and regression method. The effects of varying nonresponse rates were also investigated. The mean deviation, mean absolute deviation, bias, and the root mean square deviation were the criteria used in determining the better multiple imputation method. Results showed that the predictive mean matching method is better than the regression method.
Abstract Format
html
Language
English
Format
Accession Number
TU15996
Shelf Location
Archives, The Learning Commons, 12F, Henry Sy Sr. Hall
Physical Description
1 v. (various foliations) ; ill. ; 28 cm.
Keywords
Multiple imputation (Statistics); Statistical matching; Regression analysis; Cost and standard of living--Philippines--Statistics
Recommended Citation
Hao, M. (2010). Comparison of predictive mean matching and regression multiple imputation methods using the 2006 Family Income and Expenditure Survey (FIES). Retrieved from https://animorepository.dlsu.edu.ph/etd_bachelors/2443
