Suppose I have two time series (Series1 and Series2) which are identical from timePeriod -200 to timePeriod 0. Say that in timePeriod 0 they are both equal to 100. Series1is equal to 200 at timePeriod 50. Series2 is equal to 250 at timePeriod 50. They both go up and down over time, but Series2 is always equal to Series1+timePeriod number. (Ex. at timePeriod 20, it equals Series1+20.)

Obviously there is a significant difference between the levels of these 2 series at timePeriod 50. However, is there a way I can test to see whether there is a (statistically) significant difference in the trends of these two series over time? (Preferably, the method could be used with different numbers and time periods)


This is a useful text on "Abrupt change detection" by Michèle Basseville and Igor V. Nikiforov,. I think it can answer the substance of your question. I think the CUSUM methods might do well there.


I am not sure about the method you describe. After some thought, it sounds like you are asking about a correspondence between the two. If so then making a lag plot or a scatter plot can be useful.

  • $\begingroup$ Thanks, looks pretty complicated, I'll run it by some people who might be able to understand it better. $\endgroup$ – catquas May 27 '13 at 19:07

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.