Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование
Bahadur efficiency and local optimality of a test for the exponential distribution based on the gini statistic. / Nikitin, Ya Yu; Tchirina, A. V.
в: Statistical Methods and Applications, Том 5, № 1, 01.01.1996, стр. 163-175.Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование
}
TY - JOUR
T1 - Bahadur efficiency and local optimality of a test for the exponential distribution based on the gini statistic
AU - Nikitin, Ya Yu
AU - Tchirina, A. V.
PY - 1996/1/1
Y1 - 1996/1/1
N2 - The sample scale-free Gini index is known to be a powerful test of exponentiality against a broad class of alternatives. To understand better the efficiency properties of this test we calculate its Bahadur efficiency for most commonly used parametric alternatives to the exponential distribution. Using variational arguments and the Bahadur-Raghavachari inequality for exact slopes we find the conditions of local Bahadur optimality of the Gini test. It turns out that this property surprisingly holds for a family of alternative distributions including the well-known GompertzMakeham distribution.
AB - The sample scale-free Gini index is known to be a powerful test of exponentiality against a broad class of alternatives. To understand better the efficiency properties of this test we calculate its Bahadur efficiency for most commonly used parametric alternatives to the exponential distribution. Using variational arguments and the Bahadur-Raghavachari inequality for exact slopes we find the conditions of local Bahadur optimality of the Gini test. It turns out that this property surprisingly holds for a family of alternative distributions including the well-known GompertzMakeham distribution.
UR - http://www.scopus.com/inward/record.url?scp=0039791765&partnerID=8YFLogxK
U2 - 10.1007/BF02589587
DO - 10.1007/BF02589587
M3 - Article
AN - SCOPUS:0039791765
VL - 5
SP - 163
EP - 175
JO - Statistical Methods and Applications
JF - Statistical Methods and Applications
SN - 1618-2510
IS - 1
ER -
ID: 47771509