I have a question on a ACF/PACF result reported below.

Using a daily time series composed by 953 values I have the following result plotting the acf and pacf results.

enter image description here

My questions are:

  1. how can I interpret the fact that in the ACF there are a kind of pick at lag 7, 14, 21 ,ecc ? It seems drop down only at 600th lag

  2. how can I interpret the PACF result due to this stange behavior ?


This would indicate a clear seasonal effect of period 7 (such as you might see with a "day-of-week" type effect and daily data)

The fact that it extends for a long time suggests you might need seasonal differencing.

If you look at the plots for seasonally differenced series you'll probably see indication of a small ARMA model -- perhaps an MA(1), say.


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.