SOME HISTORICAL REFLECTIONS TRACED THROUGH THE DEVELOPMENT OF THE USE OF FREQUENCY CURVES.
Abstract
The paper is mainly concerned with the uses made of Karl Pearson's system of frequency curves, in particular with developments originating in the Department of Statistics at University College London. After a brief account of the history of the Normal curve from the time of DeMoivre to Francis Galton, Pearson's derivation of his system of curves from a single differential equation is described. The system was originally intended for the gradation of observed frequency distributions which the Normal curve would not fit. But since Student's pioneer work on the distributions of s squared, t and r (when rho = 0) the curves have been used again and again to approximate the sampling distributions of statistics whose moments only are known. They have also been used in Monte Carlo experiments to examine the distribution of statistics based on samples taken from representative non-normal populations. The report illustrates some of this work and discusses conclusions to be drawn from it. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 03, 1969
- Accession Number
- AD0693042
Entities
People
- Egon S. Pearson
Organizations
- Southern Methodist University