Confidence distribution inferences in one-way random effects model

Xuhua Liu ^[1] ; Xingzhong Xu ^[2]
1. [1] China Agricultural University
  
  China Agricultural University
  
  China
2. [2] Beijing Institute of Technology
  
  Beijing Institute of Technology
  
  China
Localización: Test: An Official Journal of the Spanish Society of Statistics and Operations Research, ISSN-e 1863-8260, ISSN 1133-0686, Vol. 25, Nº. 1, 2016, págs. 59-74
Idioma: inglés
DOI: 10.1007/s11749-015-0440-8
Texto completo no disponible (Saber más ...)
Resumen
- In this paper, we construct a new kind of confidence intervals for the parameters of interest through constructing confidence distributions for them in one-way random effects model. At first, we use the method of Singh et al. (Ann Stat 33:159–183, 2005) to derive combined asymptotic confidence distribution, then obtain the confidence interval naturally by the property of confidence distribution. Simulation results demonstrate that the new confidence intervals perform very well in terms of empirical coverage probability and average interval length. Although we focus on confidence interval estimation, our method can also be used to carry out hypothesis tests about the parameters of interest.