A 0-1 quadratic programme for the case of missing data in regression. (1st January 2014)

• Record Type: Journal Article
• Authors: Smith, Brian K.; Chimka, Justin R.; Nachtmann, Heather
• Is Part Of: International journal of data analysis techniques and strategies. Volume 6:Number 1(2014)
• Page Start: 94
• Page End: 104
• Publication Date: 2014-01-01
• Subjects: missing data -- quadratic programme -- QP -- regression analysis -- survey research; Electronic data processing -- Periodicals; Database searching -- Periodicals; 005
• Languages: English
• ISSNs: 1755-8050

Abstract

Multivariate statistical analysis techniques including regression analysis compose a popular toolset for analysing survey data, but the techniques require a complete dataset with no missing values. Unfortunately, most survey datasets contain missing values. These missing values must be resolved in some manner before regression analysis can take place. We present a quadratic programming methodology for eliminating non-responses from a survey dataset.