抄録
We develop Granger causality tests that apply directly to data sampled at different frequencies. We show that taking advantage of mixed frequency data allows us to better recover causal relationships when compared to the conventional common low frequency approach. We also show that the new causality tests have higher local asymptotic power as well as more power in finite samples compared to conventional tests. In an empirical application involving U.S. macroeconomic indicators, we show that the mixed frequency approach and the low frequency approach produce very different causal implications, with the former yielding more intuitively appealing result.
| 元の言語 | English |
|---|---|
| ページ（範囲） | 207-230 |
| ページ数 | 24 |
| ジャーナル | Journal of Econometrics |
| 巻 | 192 |
| 発行部数 | 1 |
| DOI | |
| 出版物ステータス | Published - 2016 5 1 |
Fingerprint
ASJC Scopus subject areas
- Economics and Econometrics
- Applied Mathematics
- History and Philosophy of Science
これを引用
Testing for Granger causality with mixed frequency data. / Ghysels, Eric; Hill, Jonathan B.; Motegi, Kaiji.
：: Journal of Econometrics, 巻 192, 番号 1, 01.05.2016, p. 207-230.研究成果: Article
}
TY - JOUR
T1 - Testing for Granger causality with mixed frequency data
AU - Ghysels, Eric
AU - Hill, Jonathan B.
AU - Motegi, Kaiji
PY - 2016/5/1
Y1 - 2016/5/1
N2 - We develop Granger causality tests that apply directly to data sampled at different frequencies. We show that taking advantage of mixed frequency data allows us to better recover causal relationships when compared to the conventional common low frequency approach. We also show that the new causality tests have higher local asymptotic power as well as more power in finite samples compared to conventional tests. In an empirical application involving U.S. macroeconomic indicators, we show that the mixed frequency approach and the low frequency approach produce very different causal implications, with the former yielding more intuitively appealing result.
AB - We develop Granger causality tests that apply directly to data sampled at different frequencies. We show that taking advantage of mixed frequency data allows us to better recover causal relationships when compared to the conventional common low frequency approach. We also show that the new causality tests have higher local asymptotic power as well as more power in finite samples compared to conventional tests. In an empirical application involving U.S. macroeconomic indicators, we show that the mixed frequency approach and the low frequency approach produce very different causal implications, with the former yielding more intuitively appealing result.
KW - Granger causality test
KW - Local asymptotic power
KW - Mixed data sampling (MIDAS)
KW - Temporal aggregation
KW - Vector autoregression (VAR)
UR - http://www.scopus.com/inward/record.url?scp=84960146256&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84960146256&partnerID=8YFLogxK
U2 - 10.1016/j.jeconom.2015.07.007
DO - 10.1016/j.jeconom.2015.07.007
M3 - Article
AN - SCOPUS:84960146256
VL - 192
SP - 207
EP - 230
JO - Journal of Econometrics
JF - Journal of Econometrics
SN - 0304-4076
IS - 1
ER -