Exponential means-singly censored. Suppose that one observes the first rk order statistics of a sample of size n , from an exponential distribution with mean θk = 1,. . . , K . where the samples are statistically independent.
(a) Derive the test statistic of the likelihood ratio test for equality
(b) Use this test on the data in Table 2.1 of Chapter 7. Determine a simple approximate and satisfactory way of handling left and multiply censored samples there.
(c) Do you think that the asymptotic theory is adequate for this application? Explain why.
(d) Plot the samples on Weibull paper. Is a formal hypothesis test needed?
Table 2.1
