I am new to recurrence plots and RQA, but I have read some of the papers Norbert recommended. Still, I have trouble figuring out how to caracterize which system leads the other one with two given time series.
I am trying to analyse 4 binary vectors corresponding to talk turns in a meeting and to extract dominance from those info.
So far, I have tried to find the optimal parameters to use with crp and crqad (embedding dimension, delay and threshold). Using pss, fnn and mi, I find threshold=0.1, embdim=11 and delay=1.
I then use the commands crp and crqad with these values and the option 'fan' between two of my time series.
I have obtained the following results, but I am not sure how to analyse them.
CRP :
CRQA :
I think something is wrong with my results (bad choice of parameters ?), and I am also not sure how to extract information about dominance from it...
Thanks for your help,
Dalayrac.