A COMPUTER PROGRAM FOR THE KOLMOGOROV GOODNESS OF FIT TEST FOR NORMALITY

Abstract

Following a brief discussion on the background, applicability, and limitations of the Kolmogorov Test, a computer program of the Kolmogorov Test for normality is described. The program is applicable for testing the null hypothesis that a random sample is from a parent normal population whose mean and variance are equal to those of the sample distribution. The program determines the minimum and maximum sample values, computes sample estimates of the mean and variance of the hypothesized normal distribution, and computes the Kolmogorov statistic. Five optional significance levels (0.20, 0.15, 0.10, 0. 05, and 0.01) are available, and sample size limitations are 4 < or = n < or = 2500. An optional feature provides for CRT plot output applicable for both hypothesis testing and interval estimation. The program is coded in FORTRAN IV for the IBM 7030 (STRETCH) computer.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1968
Accession Number
AD0665291

Entities

People

  • Carl B. Bates
  • Jacqueline R. Orsulak

Organizations

  • Naval Surface Warfare Center Dahlgren Division

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Chi Square Test
  • Computer Programming
  • Computer Programs
  • Computers
  • Confidence Limits
  • Data Science
  • Distribution Functions
  • Information Science
  • Normal Distribution
  • Normality
  • Statistical Analysis
  • Statistical Samples
  • Statistics

Fields of Study

  • Mathematics

Readers

  • Computer Science.
  • Statistical inference.