Comparison of the SAS, SPSS and STATISTICA systems in the area of clustering variables

  • Authors:

  • Hana Rezanková;Dušan Húsek

  • Affiliations:

  • Department of Statistics and Probability, University of Economics, Prague 130 67 Prague, Czech Republic;Institute of Computer Sciences, Academy of Sciences of the Czech Republic 182 00 Prague, Czech Republic

  • Venue:
  • Computational Statistics & Data Analysis
  • Year:
  • 2002

Quantified Score

Hi-index 0.03

Visualization

Abstract

Techniques for clustering variables that are contained in statistical packages are characterized, and capabilities of the SAS, SPSS and STATISTICA systems in this area are shown. The paper concentrates on factor analysis, nonlinear principle component analysis, cluster analysis, multidimensional scaling and multiple correspondence analysis. Important capabilities of the S-PLUS package and STATISTICA Neural Network software product are described. The main restrictions for the use of some methods are mentioned.