Transformation of the Pearson System with Special Reference to Type IV
The Pearson system of distributions consists of several types, including beta, gamma, and F-ratio families. There are interrelations between types, resulting from non-linear transformation of the variates. A description of these interrelations is given along with a new transformation which carries Type IV family into distributions for which (√ß1, ß2) points lie in Type I region. The usefulness of the transformation is illustrated in an application where the percentage points of the moment estimator for the shape parameter of the gamma distribution are derived.
Key WordsTransformation Pearson system of curves
Unable to display preview. Download preview PDF.
- Dusenberry, W. E. and Bowman, K. O. (1973). ORNL Report No. 4876, Oak Ridge National Laboratory, Oak Ridge, Tennessee.Google Scholar
- Johnson, N. L. and Kotz, S. (1970). Distributions in Statistics, Continuous Univariate Distributions - 1. Houghton Mifflin, Boston, and John Wiley.Google Scholar
- Pearson, E. S. (1963). Biometrika 50, 1 and 2, 95–112.Google Scholar
- Shenton, L. R. (1965). Georgia Ag. Exp. Sta. Tech. Bull. N.S. 50, pp. 1–48.Google Scholar
- Shenton, L. R. and Bowman, K. O. (1975). Johnson’s SU and the skewness and kurtosis statistics. Jour. Amer. Statist. Assoc. (to appear).Google Scholar