IGARCH effect on autoregressive lag length selection and causality tests
Using Monte Carlo experiments, we show how information criteria determine, in the presence of GARCH errors, an optimal lag length in univariate time series and causality tests. We illustrate the simulations by testing the presence of serial correlation in exchange rates as well as Granger-causality between interest rates.
Year of publication: |
1996
|
---|---|
Authors: | Hecq, Alain |
Published in: |
Applied Economics Letters. - Taylor & Francis Journals, ISSN 1350-4851. - Vol. 3.1996, 5, p. 317-323
|
Publisher: |
Taylor & Francis Journals |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Testing for Common Cyclical Features in Nonstationary Panel Data Models
Hecq, Alain, (2000)
-
Testing for Common Cyclical Features in Var Models with Cointegration
Hecq, Alain, (2001)
-
Separation, Weak Exogeneity and P-T Decomposition in Cointegrated VAR Systems with Common Features
Hecq, Alain, (2002)
- More ...