Distributions of exact tests in the exponential family

被引：29
|
作者：
Stehlik, M [1 ]
机构：
[1] Comenius Univ, Fac Math Phys & Informat, Dept Probabil & Stat, Bratislava 84248, Slovakia
来源：
METRIKA | 2003年 / 57卷 / 02期
关键词：
exponential family; I-divergence; model testing; maximum likelihood; asymptotic efficiency; exact slopes; generalized linear models; Lambert W function;
D O I：
10.1007/s001840200206
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
The aim of this paper is to give some results on the exact density of the I-divergence in the exponential family with gamma distributed observations. It is shown in particular that the I-divergence can be decomposed as a sum of two independent variables with known distributions. Since the considered I-divergence is related to the likelihood ratio statistics, we apply the method to compute the exact distribution of the likelihood ratio tests and discuss the optimality of such exact tests. One of these tests is the exact LR test of the model which is asymptotically optimal in the Bahadur sense. Numerical examples are provided to illustrate the methods discussed.
引用
收藏
页码：145 / 164
页数：20
相关论文
共 50 条
  • [41] On empirical Bayes tests in a positive exponential family
    Liang, TC
    JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2000, 83 (01) : 169 - 181
  • [42] Exact One- and Two-Sample Likelihood Ratio Tests based on Time-Constrained Life-Tests from Exponential Distributions
    Xiaojun Zhu
    Narayanaswamy Balakrishnan
    Hon-Yiu So
    Methodology and Computing in Applied Probability, 2022, 24 : 2009 - 2028
  • [43] Exact One- and Two-Sample Likelihood Ratio Tests based on Time-Constrained Life-Tests from Exponential Distributions
    Zhu, Xiaojun
    Balakrishnan, Narayanaswamy
    So, Hon-Yiu
    METHODOLOGY AND COMPUTING IN APPLIED PROBABILITY, 2022, 24 (03) : 2009 - 2028
  • [44] UMVU estimation for certain family of exponential distributions.
    LopezBlazquez, F
    BarrancoChamorro, I
    MorenoRebollo, JL
    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1997, 26 (02) : 469 - 482
  • [45] q-exponential fitting for distributions of family names
    Yamada, Hiroaki S.
    Iguchi, Kazumoto
    PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS, 2008, 387 (07) : 1628 - 1636
  • [46] A Computationally Efficient Method for Learning Exponential Family Distributions
    Shah, Abhin
    Shah, Devavrat
    Wornell, GregoryW.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [47] Mixture of Distributions in the Biparametric Exponential Family: A Bayesian Approach
    Liliana Garrido, L.
    Cepeda, Edilberto
    COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2012, 41 (03) : 355 - 375
  • [48] Graphical Models via Univariate Exponential Family Distributions
    Yang, Eunho
    Ravikumar, Pradeep
    Allen, Genevera I.
    Liu, Zhandong
    JOURNAL OF MACHINE LEARNING RESEARCH, 2015, 16 : 3813 - 3847
  • [49] Model selection for the segmentation of multiparameter exponential family distributions
    Cleynen, Alice
    Lebarbier, Emilie
    ELECTRONIC JOURNAL OF STATISTICS, 2017, 11 (01): : 800 - 842
  • [50] On Exponential Negative-Binomial-X Family of Distributions
    Hussain Z.
    Aslam M.
    Asghar Z.
    Annals of Data Science, 2019, 6 (04) : 651 - 672
12345