On the kolmogorov-smirnov test for normality with mean and variance unknown

Disponível somente no TrabalhosFeitos
  • Páginas : 7 (1750 palavras )
  • Download(s) : 0
  • Publicado : 13 de abril de 2013
Ler documento completo
Amostra do texto
On the Kolmogorov-Smirnov Test for Normality with Mean and Variance Unknown Hubert W. Lilliefors Journal of the American Statistical Association, Vol. 62, No. 318. (Jun., 1967), pp. 399-402.
Stable URL: http://links.jstor.org/sici?sici=0162-1459%28196706%2962%3A318%3C399%3AOTKTFN%3E2.0.CO%3B2-G Journal of the American Statistical Association is currently published by American StatisticalAssociation.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for yourpersonal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/astata.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

The JSTOR Archive is a trusted digital repositoryproviding for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR,please contact support@jstor.org.

http://www.jstor.org Tue Jan 29 18:38:16 2008


HUBERT LILLIEFORS W. The George Washington University
The standard tables used for the Kolmogorov-Smirnov test are valid when testing whether a set of observations are from a completely specified continuous distribution. If oneor more parameters must be estimated from the sample then the tables are no longer valid. A table is given in this note for use with the Kolmogorov-Smirnov statistic for testing whether a set of observations is from a normal population when the mean and variance are not specified but must be estimated from the sample. The table is obtained from a Monte Carlo calculation. A brief Monte Carloinvestigation is made of the power of the test.

Kolmogorov-Smirnov statistic provides a means of testing whether a set of observations are from some completely specified continuous distribution, F o ( X ) . The usual alternative would be the chi-square test. The Kolmogorov-Smirnov test has a t least two major advantages over the chisquare test [ref. 1, 21.



1. It can be used with smallsample sizes, where the validity of the chisquare test would be questionable. 2. Often it appears to be a more powerful test than the chi-square test for any sample size. Unfortunately, when certain parameters of the distribution niust be estimated from the sample, then the Kolmogorov-Smirnov test no longer appliesat least not using the commonly tabulated critical points. I t is suggested in ref. 2that if the test is used in this case, the results will be conservative in the sense that the probability of a type I error will be smaller than as given by tables of the Kolmogorov-Smirnoo statistic [as found in ref. 2 or 41. As will be seen below, the results of this procedure will indeed be extremely conservative. In ref. 1 it is pointed out that if the parameters to be estimated areparameters of scale or location, then one can construct tables for use with the Kolmogorov-Smirnov statistic for that particular distribution. This note presents a table for use with the Kolmogorov-Smirnov statistic when testing that a set of observations are from a normal population but the mean and variance are not specified. The procedure is: Given a sample of N observations, one determines D =maxx /...
tracking img