MIXED NORMAL INFERENCE ON MULTICOINTEGRATION
Asymptotic likelihood analysis of cointegration in <italic>I</italic> (2) models (see Johansen, 1997, 2006; Boswijk, 2000; Paruolo, 2000) has shown that inference on most parameters is mixed normal, implying hypothesis test statistics with an asymptotic <italic>χ</italic><sup>2</sup> null distribution. The asymptotic distribution of the multicointegration parameter estimator so far has been characterized by a Brownian motion functional, which has been conjectured to have a mixed normal distribution, based on simulations. The present note proves this conjecture.
Year of publication: |
2010
|
---|---|
Authors: | Boswijk, H. Peter |
Published in: |
Econometric Theory. - Cambridge University Press. - Vol. 26.2010, 05, p. 1565-1576
|
Publisher: |
Cambridge University Press |
Description of contents: | Abstract [journals.cambridge.org] |
Saved in:
Saved in favorites
Similar items by person
-
Cointegration, identification and exogeneity : inference in structural error correction models
Boswijk, H. Peter, (1992)
-
Identifiability of cointegrated systems
Boswijk, Herman Peter, (1995)
-
Mixed normality and ancillarity in (2) systems
Boswijk, Herman Peter, (1996)
- More ...