View : 546 Download: 0

Tests for double unit roots based on symmetric estimators and recursive mean adjustment

Title
Tests for double unit roots based on symmetric estimators and recursive mean adjustment
Authors
이주희
Issue Date
2000
Department/Major
대학원 통계학과
Publisher
이화여자대학교 대학원
Degree
Master
Abstract
우리는 평균이 0이고 분산이 σ ²인 error process에 대한 모형에서 단위근이 2개인지를 검정하는 새로운 검정법들을 개발하였다. 새로운 검정법들은 recursive 평균수정방법과 symmetric 추정법에 근거하고 있다. 몬테카를로 시뮬레이션을 통해서 새 검정법들을 기존의 Sen and Dickey의 symmetric 검정법과 비교한다. 새 검정법은 둘다 명목상의 size에 근접한 empirical size를 가진다. 게다가 그 제안된 검정법들은 symmetric 검정법보다 power가 더 좋다. 특히 symmetric 추정법과 recursive 평균수정법을 모두 적용한 검정법은 symmetric 추정법이나 recursive 평균수정법 하나에만 근거한 검정법보다 월등하게 power가 높다. ; We develop new tests for double unit roots of models with error process having zero mean and variance σ². The new tests are based on the recursive mean adjustment method and the symmetric estimation. In a Monte Carlo Simulation, the new tests are compared with the previously existing symmetric test of Sen and Dickey. Both of the new two tests have empirical sizes close to the nominal size. Moreover, the proposed tests have better power performance than the symmetric test. Especially, the test based on symmetric estimation and recursive mean adjustment simultaneously has substantially better power than other tests based only one of symmetric estimation or recursive mean adjustment.
Fulltext
Show the fulltext
Appears in Collections:
일반대학원 > 통계학과 > Theses_Master
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE