Hausman Principal Component Analysis
- 1.6k Downloads
The aim of this paper is to obtain discrete-valued weights of the variables by constraining them to Hausman weights (−1, 0, 1) in principal component analysis. And this is done in two steps: First, we start with the centroid method, which produces the most restricted optimal weights −1 and 1; then extend the weights to −1,0 or 1.
KeywordsPrincipal Component Analysis Centroid Method Multiple Factor Analysis Ordinary Principal Component Analysis Principal Component Analysis Loading
Unable to display preview. Download preview PDF.
- BURT, C. (1917): The Distribution And Relations Of Educational Abilities. P.S. King & Son: London.Google Scholar
- CHIPMAN, H.A., and GU, H. (2003): Interpretable dimension reduction. To appear in Journal of Applied Statistics.Google Scholar
- CHOULAKIAN, V. (2003): The optimality of the centroid method. Psychometriks, 68, 473–475.Google Scholar
- CHOULAKIAN, V. (2005b): L1-norm projection pursuit principal component analysis. Computational Statistics and Data Analysis, in press.Google Scholar
- JACKSON, J.E. (1991): A User’s Guide To Principal Components. Wiley: New York.Google Scholar
- JOLLIFFE, I.T. (2002): Principal Component Analysis. Springer Verlag: New York, 2nd edition.Google Scholar
- THURSTONE, L.L. (1931): Multiple factor analysis. Psychological Review, 38, 406–427.Google Scholar
- WOLD, H. (1966): Estimation of principal components and related models by iterative least squares. In Krishnaiah, P.R., ed.: Multivariate Analysis, Academic Press, N.Y., pp. 391–420.Google Scholar