The Power of McCabe and Tremayne Test (1995) for Difference Stationarity When the Errors are Correlated
| Autor: | I-Po Chen, 陳易伯 |
|---|---|
| Rok vydání: | 2018 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 107 In this thesis, I relax McCabe and Smith (1998)’s assumptions to reconsider the power properties of McCabe and Tremayne (1995, MT) test for the difference stationarity of a times series. Without the independence assumption between the random coefficient and error process as that of McCabe and Smith (1998), I derive the asymptotic distribution of MT test under local heteroscedastic integration alternative. I find that the MT test statistics is O(T1/4) and therefore it is diverge. In a finite sample, the power increase as the correlation or the variance of error process increase. Monte Carlo evidence supports our theoretical findings. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|