Universal Journal of Mathematics and Mathematical Sciences
Volumes 7, Issue 1, Pages 1 - 15
(January 2015) http://dx.doi.org/10.17654/UJMMSJan2015_001_015 |
|
TESTING EXPONENTIAL BETTER THAN USED IN CONVEX AVERAGE CLASS OF LIFE DISTRIBUTIONS BASED ON KERNEL METHODS
N. M. Al-Ghufily
|
Abstract: A new test for testing that a life distribution is exponential against the alternative that is exponential better (worse) than used in convex average EBUCA (EWUCA), but not exponential is presented based on the highly popular “Kernel methods” of curve fitting. This new procedure is competitive with the old one in the sense of Pitman’s asymptotic relative efficiency. The test is easy to compute and it is not dependent on the bandwidth or kernel. It also has a good power. |
Keywords and phrases: test exponentiality, life distribution classes, Kernel method, asymptotic normality, efficiency, test power, EBUCA. |
|
Number of Downloads: 277 | Number of Views: 1032 |
|