0
$\begingroup$

Could someone give me some suggestions as to how I can go about trying to detect if a time series is sporadic or not? Are there any tests for the same? Also, I think sporadic series are quite different from random or white noise series and therefore the tests like Ljung box tests do not work.

Could someone help me with some ideas to test for sporadic series?

$\endgroup$
2
  • 2
    $\begingroup$ How are you defining "sporadic"? $\endgroup$
    – jbowman
    Aug 16, 2021 at 17:23
  • $\begingroup$ I mean something like an Intermittent time series where the peaks occur occasionally. $\endgroup$ Aug 17, 2021 at 7:31

1 Answer 1

0
$\begingroup$

I'm not clear on what kind of time series your testing for when you say sporadic, but one popular option for testing if a time series is composed of time-dependent values is the Augmented Dickey-Fuller Test.

$H_0: \text{The series is not stationary (series values are time dependent)}$
$H_A: \text{The series is stationary, evidence of no trend (series values are not time dependent)}$

A stationary time series is a time series:

whose properties do not depend on the time at which the series is observed.

https://otexts.com/fpp2/stationarity.html

The site above is a good introduction or review resource for stationarity in time series.

$\endgroup$

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.