Advances and Applications in Statistics
Volume 60, Issue 2, Pages 201 - 215
(February 2020) http://dx.doi.org/10.17654/AS060020201 |
|
MODIFIED JACKKNIFE EMPIRICAL LIKELIHOOD
Ramadha D. Piyadi Gamage, Ying-Ju Chen and Wei Ning
|
Abstract: Jackknife empirical likelihood (JEL) is an effective modified version of empirical likelihood method (EL). Through the construction of the jackknife pseudo-values, JEL overcomes the computational difficulty of EL method when its constraints are nonlinear while maintaining the same asymptotic results for one sample and two-sample U statistics. In this paper, we propose an adjusted version of JEL to guarantee that the adjusted jackknife empirical likelihood (AJEL) statistic is well-defined for all the values of the parameter, instead of restricting on the convex hull of the estimation equation. The properties of JEL have been preserved for AJEL. |
Keywords and phrases: jackknife empirical likelihood, adjusted empirical likelihood, coverage probability, confidence interval, Wilks’ theorem.
|
|
Number of Downloads: 326 | Number of Views: 906 |
|