In most multivariate statistical procedures used in the field of Education and Psychology, the n by n matrix of intercorrelations among the n attributes or variables is required to be a symmetric positive definite matrix, i.e., an non-singular matrix. When a matrix is close to being singular it is said to be "ill-conditioned." In the present note, the triangular decomposition method is suggested as a general technique for obtaining the various measures of an ill-conditioned matrix.
Get full access to this article
View all access options for this article.
References
1.
Berk, K. N.Tolerance and condition in regression computations. In D. Hoaglin and R. Welsch (Eds.), Proceedings of the ninth interface symposium on computer science and statistics. Boston: Prindle, Weber, and Schmidt, 1976.
2.
Dixon, W. J.BMD Biomedical Computer Programs. Berkeley: University of California Press, 1973.
3.
Dixon, W. J.BMDP Biomedical Computer Programs. Berkeley: University of California Press, 1977.
4.
Dwyer, P. S.The square root method and its use in correlation and regression . Journal of American Statistical Association, 1945, 40, 493-503.
5.
Dwyer, P. S.Linear computations. New York: Wiley, 1951.
6.
Efroymson, M. A.Multiple regression analysis. In A. Ralston and H. Wilf (Eds.), Mathematical methods for digital computers. New York: Wiley, 1960.
7.
Fox, L.Introduction to numerical linear algebra. London: Oxford University Press, 1964.
8.
Frane, J. W.A note on checking tolerance in matrix inversion and regression . Technometrics, 1977, 19, 513-514.
9.
Harman, H. H.Modern factor analysis (3rd ed.). Chicago: University of Chicago Press, 1976.
10.
Holzinger, K. J.Preliminary reports on Spearman-Holzinger unitary trait study, nos. 1-9. Chicago: University of Chicago, Statistics Laboratory, Department of Education, 1936.
11.
Longley, J. W.An appraisal of least squares programs for the electronic computer from the point of view of the user . Journal of American Statistical Association, 1967, 62, 819-841.
12.
Marquardt, D. W.Generalized inverses, ridge regression, biased linear estimation, and nonlinear regression . Technometrics, 1970, 12, 591-613.
13.
Nie, N. H. , Hull, C. H., Jenkins, J., Steinbrenner, K., and Bent, D.Statistical Package for the Social Sciences (2nd ed.). New York: McGraw-Hill, 1975.
14.
Pinneau, S. R. and Ault, J. T.Univariate and multivariate component analysis. Perceptual and Motor Skills, 1974, 39, 955-985.
15.
Pinneau, S. R. and Ender, P. B.Analysis of component variance. Perceptual and Motor Skills, 1971, 33, 327-335.
16.
Tatsuoka, M. M.Multivariate analysis: Technique for educational and psychological research. New York: Wiley, 1971.
17.
Thurstone, L. L.The vectors of mind. Chicago: University of Chicago Press, 1935.
18.
Young, D. M. and Gregory, R. T.A survey of numerical mathematics. Reading: Addison-Wesley, 1973.
