Goodness-of-fit tests based on Kullback-Leibler information


ŞENOĞLU B., Surucu B.

IEEE TRANSACTIONS ON RELIABILITY, vol.53, no.3, pp.357-361, 2004 (SCI-Expanded, Scopus)

  • Publication Type: Article / Article
  • Volume: 53 Issue: 3
  • Publication Date: 2004
  • Doi Number: 10.1109/tr.2004.833319
  • Journal Name: IEEE TRANSACTIONS ON RELIABILITY
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.357-361
  • Keywords: BLUE, correlation, distribution, entropy, exponential, kurtosis, normal, skewness, spacings, uniform, ENTROPY-BASED TESTS, SUSPECTED OUTLIERS, COMPLETE SAMPLES, EMPIRICAL POWER, VARIANCE TEST, EXPONENTIALITY, NORMALITY, DISTRIBUTIONS, STATISTICS, PLOT
  • Ankara University Affiliated: No

Abstract

We evaluate the power of the sample entropy goodness-of-fit tests for s-normal, exponential, and uniform distributions. We compare them with the mainstream statistical tests, the W test based on the best linear unbiased estimator (BLUE) of the location parameter, the Z test based on the sample spacings, and the R test based on the correlation coefficient between the order statistics of the sample & the corresponding population quantiles. We show that the latter are more powerful overall. The, mainstream statistical tests, particularly the Z test, readily extend to censored samples and to multi-sample situations.