A men goodness-of-fit test for the two-parameter wetbull or extreme-value distribution with unknown parameters

Abstract
A new test of fit to the two-parameter Weibull or extreme-value distribution with unknown parameters is developed in this paper. This test is known to have desirable power properties, relative to analogues of certain "classical" tests, against two important classes of alternatives. The test statistic is easy to calculate and can be used for censored samples. Percentage points and certain expected values which are needed to implement the test are provided for sampies of size n = 3(1)25. the closeness of these cercentage points, aven for small n, to Beta pertentage points and the existence of tables of the needed expected values for n = 1(1) 60(5)100 make it possible to use the test in the range 25 < n lt; 1as well