0
$\begingroup$

I'm using KPSS method to check if the time series is stationary, but I would also like to use another test to confirm the hypothesis that the time series is stationary.

I'm already using Unit root tests as ADF or PP, but I need a more specific test to check the stationarity.

What method should I use?

$\endgroup$
2
$\begingroup$

KPSS, ADF and PP are a good way to start the analysis. The main problem of ADF and PP is that they have low power against I(0) alternatives that are close to I(1). To overcome this problem some test have been developed such as Elliot, Rothenberg and Stock.

Zivot has a very nice discussion about unit root testing at: http://faculty.washington.edu/ezivot/econ584/notes/unitroot.pdf

$\endgroup$
  • $\begingroup$ I'm using KPSS, ADF, PP and ERS unit root test, So I think i'm using all the test about unit roots, the problem is that the series have not unit root, ok, but have strange different volatility during all the period... maybe an homoscedastic test to filter the series that have different variance(residuals) ? $\endgroup$ – Dail Nov 13 '11 at 17:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.