I am trying to analyze these pltos but having a hard time since I don't know if they are even right... enter image description here

enter image description here

So i follow your advice and these new plots is what i got: enter image description here enter image description here


So the ACF plot tests if there is a significant correlation between X and the lags of X. So in your case, it seems that you have a very high correlation between X and the 1st lag of X. The blue cone represents the level of significance you need for it to be a significant correlation. As for the PACF, that says, if I make an linear model out of all of the previous lags, is there still any significant lags left?

So in this case it looks like you have a high correlation of around 0.9 and X_t and X_{t-1} and the PACF says that once you add to your model, lags 2-7 will not needed to be added. I recommend adding lag of 1 to your model and looking at the acf/pacf plots again.

| cite | improve this answer | |
  • $\begingroup$ Hey! So I followed your advice and added a lag of 1 to my model and got these new acf and pacf plots, would you mind interprating them for me to make sure I'm understanding? (New plots on question) $\endgroup$ – Fernanda May 4 at 20:17
  • $\begingroup$ Great. Your model should probably look better now.(I hope) what I would do next is add the next big spike, which is lag 2 I think $\endgroup$ – Anonymous Emu May 4 at 21:07

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