asad009.pdf (719.61 kB)
Sampling distribution for single-regression Granger causality estimators
Version 2 2023-07-20, 12:24
Version 1 2023-06-10, 06:15
journal contribution
posted on 2023-07-20, 12:24 authored by AJ Gutknecht, Lionel BarnettLionel BarnettThe single-regression Granger-Geweke causality estimator has previously been shown to solve known problems associated with the more conventional likelihood-ratio estimator; however, its sampling distribution has remained unknown. We show that, under the null hypothesis of vanishing Granger causality, the single-regression estimator converges to a generalized ?2 distribution, which is well approximated by a G distribution. We show that this holds too for Geweke's spectral causality averaged over a given frequency band, and derive explicit expressions for the generalized ?2 and G-approximation parameters in both cases. We present a Neyman–Pearson test based on the single-regression estimators, and discuss how it may be deployed in empirical scenarios. We outline how our analysis may be extended to the conditional case, point-frequency spectral Granger causality, and the important case of state-space Granger causality.
History
Publication status
- Published
File Version
- Published version
Journal
BiometrikaISSN
0006-3444Publisher
Oxford University Press (OUP)External DOI
Page range
asad009Department affiliated with
- Informatics Publications
Full text available
- Yes
Peer reviewed?
- Yes
Legacy Posted Date
2023-02-16First Open Access (FOA) Date
2023-05-03First Compliant Deposit (FCD) Date
2023-02-15Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC