Publication Type

Journal Article

Publication Date

4-2015

Abstract

A new jackknife method is introduced to remove the first order bias in unit root models. It is optimal in the sense that it minimizes the variance among all the jackknife estimators of the form considered in Phillips and Yu (2005) and Chambers and Kyriacou (2013) after the number of subsamples is selected. Simulations show that the new jackknife reduces the variance of that of Chambers and Kyriacou by about 10% for any selected number of subsamples without compromising bias reduction. The results continue to hold true in near unit root models. (C) 2014 Elsevier B.V. All rights reserved.

Keywords

Bias reduction, Variance reduction, Jackknife, Autoregression

Discipline

Econometrics

Research Areas

Econometrics

Publication

Statistics and Probability Letters

Volume

99

First Page

135

Last Page

142

ISSN

0167-7152

Identifier

10.1016/j.spl.2014.12.014

Publisher

Elsevier

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Included in

Econometrics Commons

Share

COinS