The acf plot exponentially decays after a lag spike at 2 and the pacf plot cuts off after lag 0. So, this indicates that the time series is stationary. But after doing adf.test and Kpss test, the result tells difference story.

    Augmented Dickey-Fuller Test

data:  ltc.ts
Dickey-Fuller = 1.7982, Lag order = 3, p-value = 0.99
alternative hypothesis: stationary

P-value of adf.test is 0.99 and I can not reject the null hypothesis : non-stationary

    KPSS Test for Level Stationarity

data:  ltc.ts
KPSS Level = 0.53558, Truncation lag parameter = 1,
p-value = 0.03365

P-value of KPSS test is 0.03 and I cant reject the null hypothesis : stationary.

What would I should do in this case? Should I follow the graph or test result. Please give me an advice!

  • $\begingroup$ It is ADF test, not ACF test. Got me confused in the beginning. You could edit this. $\endgroup$ – Richard Hardy Mar 12 '18 at 9:38
  • $\begingroup$ @RichardHardy I am sorry for the typo. I have edited it. $\endgroup$ – Iamsomean Mar 12 '18 at 10:11

Your Answer

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

Browse other questions tagged or ask your own question.