Proposes a correlation-robust test statistic for coefficient randomness in local-to-unity AR models plus a nuisance-parameter-free variant that shows better power than prior tests.
In each replicatio n, we first generate {yt}T t=1 by the mechanism yt = (1 + a/T )yt− 1 + εt, t = 1, 2
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
econ.EM 1years
2023 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Testing for Coefficient Randomness in Local-to-Unity Autoregressions
Proposes a correlation-robust test statistic for coefficient randomness in local-to-unity AR models plus a nuisance-parameter-free variant that shows better power than prior tests.