I am not very good at Time Series, so I have a few questions about my ACF & PACF plots. I have read through similar questions on stack exchange and other websites, but their graphs look different. I understand somewhat how to interpret the plots, but mine do not fit the stereotypical "molds". They are not slowly decreasing, they don't seem to be significant until a certain p.

  1. For the ACF plot, because they are between the blue dotted lines, that means they are not significantly correlated?
  2. For both plots, their lags are 0.0#, is that normal? Or is that incorrect? If it is not accepted, how would I change it? Most ACF plots I see are whole numbers.
  3. For both plots, is this an AR(1) model? An ARMA model? How do I know if there is a seasonal trend? Or is this white noise because nothing is significant?
  4. Is there any other interpretation I am missing?




Your 4 questions are difficult to answer because your are presenting summary info . The ACF and PACF are descriptive/summary statistics but not always easily inferential.

Your 1000 valued time series may have either level/step shifts or be a series that needs to be differenced or transformed or segmented.

There may be some seasonality but only your data knows for sure. Post your actual data and given time and motivation some SE reader might be able to help further.

| cite | improve this answer | |

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.