Title :
k-Sample Maximum Likelihood Ratio Test for Change of Weibull Shape Parameter
Author :
Bilikam, J.Edward ; Moore, Albert H. ; Petrick, George
Author_Institution :
FMC Corporation; 1105 Coleman Ave.; PO Box 1201; San Jose, CA 95108 USA.
fDate :
4/1/1979 12:00:00 AM
Abstract :
A k-sample maximum likelihood ratio (MLR) test is derived to test equality of shape parameters for 2-parameter Weibull populations. The test is independent of the scale parameters, and the power depends on ratios of the shape parameters. Critical points and power calculations were obtained by Monte Carlo techniques for k = 2. The MLR test is equivalent to the MLR test of scale parameters for the extreme value distribution.
Keywords :
Equations; Maximum likelihood estimation; Monte Carlo methods; Reliability theory; Shape; Statistical analysis; Statistical distributions; Testing; Weibull distribution; 2-Parameter Weibull distribution; Extreme value distribution; Maximum likelihood ratio test; Model discrimination; Test of homogeneity; Test of hypothesis;
Journal_Title :
Reliability, IEEE Transactions on
DOI :
10.1109/TR.1979.5220471
