2
$\begingroup$

Say M periodic signals, then one can safely say using AR-M model can achieve the perfect prediction. But how about further, in a more general sense, is there any publications on this?

Update:

Here is my arXiv paper on this problem https://arxiv.org/abs/1902.05198

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

If you have a perfectly predictable signal periodic in frequency $f$, the pattern repeats every cycle, or $1/f$ time units.

You need a single lag to predict it, the series lagged by $1/f$ time units. So AR-$1/f$ model will achieve perfect prediction, where the only lag included is the one mentioned.

Improve this answer
$\endgroup$
2
  • $\begingroup$ yes, it is! I am now writing a paper, and I considered many cases, including the one you mentioned as a special case. But since I am not very knowledgable in the time series modeling field, I hope to see if anyone has done this. I google a lot and cannot find anything. $\endgroup$
    – ArtificiallyIntelligent
    Commented Sep 25, 2018 at 21:06
  • $\begingroup$ After half a year... Here is the arXiv! arxiv.org/abs/1902.05198 $\endgroup$
    – ArtificiallyIntelligent
    Commented Feb 19, 2019 at 20:12

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