Unification of Statistical Methods for Continuous and Discrete Data
We propose the concept of unification of statistical methods in order to develop a general philosophy of statistical data analysis. We propose that ways of thinking about statistical ends (goals) and means (procedures) are needed that provide a framework for implementing and comparing several different approaches to a data analysis problem. We believe that unification has benefits which include: existing (often parametric) methods will be better understood; many new (often nonparametric) methods will be developed. The new methods are usually computer intensive; consequently unification of statistical methods can be considered to be closely related to computational statistics. We define computational statistical methods as characterized by being graphics intensive and number crunching intensive.
KeywordsEntropy Graphite Eter Estima
Unable to display preview. Download preview PDF.
- Alexander, William (1989) “Boundary kernel estimation of the two-sample comparison density function” Texas A&M Department of Statistics Ph.D. thesis.Google Scholar
- Aly, E. A. A., M. Csorgo, and L. Horvath (1987) “P-P plots, rank processes, and Chernoff-Savage theorems” in New Perspectives in Theoretical and Applied Statistics (ed. M. L. Puri, J. P. Vilaplann, W. Wertz) New York: Wiley 135–156.Google Scholar
- Freedman, D., Pisani, R., Purves, R. (1978) Statistics, New York: Norton.Google Scholar
- Parzen, E. (1989) “Multi-sample functional statistical data analysis,” in Statistical Data Analysis and Inference, (ed. Y. Dodge). Amsterdam: North Holland, pp. 71–84.Google Scholar
- Renyi, A. (1961). “On measures of entropy and information.” Proc. 4th Berkeley Symp. Math. Statist. Probability, 1960, 1, 547–561. University of California Press: Berkeley.Google Scholar