A 0-1 quadratic programme for the case of missing data in regression

  • Authors:

  • Brian K. Smith;Justin R. Chimka;Heather Nachtmann

  • Affiliations:

  • Department of Engineering Management and Systems Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA;Department of Industrial Engineering, University of Arkansas, Fayetteville, AR 72701, USA;Department of Industrial Engineering, University of Arkansas, Fayetteville, AR 72701, USA

  • Venue:
  • International Journal of Data Analysis Techniques and Strategies
  • Year:
  • 2014

Quantified Score

Hi-index 0.00

Visualization

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.