Deterministic exponential heteroskedasticity, a weakly stationary unit-root process and a useful diagnostic test
A specific form of deterministic exponential heteroskedasticity is examined. A non-trivial unit root process which has exponentially heteroskedastic innovation and as a consequence, a variance that vanishes asymptotically is detailed. Such a unit root stochastic process, with exponential heteroskedasticity, may be perceived as weakly stationary by the usual unit root tests. In view of the importance of deterministic exponential heteroskedasticity, a new general diagnostic test for detecting the presence of deterministic exponential heteroskedasticity is developed.
Year of publication: |
2001
|
---|---|
Authors: | Vougas, Dimitrios |
Published in: |
Applied Economics Letters. - Taylor & Francis Journals, ISSN 1350-4851. - Vol. 8.2001, 6, p. 427-430
|
Publisher: |
Taylor & Francis Journals |
Saved in:
Saved in favorites
Similar items by person
-
A DIRECT TEST FOR COINTEGRATION BETWEEN APAIR OF TIME SERIES
Leybourne, Steven, (1999)
-
On the Stationarity of Futures Hedge Ratios
Degiannakis, Stavros, (2020)
-
Examining the robustness of cointegration analysis under weighted symmetric estimation
Cook, Steven, (2007)
- More ...