Page 1 of 1

CRQA with different TS lengths

Posted: Wed Feb 6, 2013 22:27
by mikecrites
Hi All,
I was wondering if there is a way to perform a CRQA with different time series lengths. I am attempting to calculate a nominal cross recurrence, which is based off of communication data. I have transcribed conversations into meaningful code (e.g., changing either words or letters into numbers). The problem is that one person is always talking more or less than the other person. To my knowledge, there is nothing in the literature to suggest a solution for such a problem.

The only way I can think of to analyze the data successfully is to transcribe the data in the way both persons spoke, code it, and run the time series on itself as a RQA. I just feel like I cannot adequately justify the results when completing the analyses that way. The two people tended to sync up their conversation toward the end, which is something I will lose with the RQA.

Anything will help on this matter. I appreciate the help.