Page 1 of 1

joint recurrence plot software

Posted: Thu Nov 1, 2007 20:18
by adrianopgpg
Someone knows if there are a software to compute the RQA measures to joint recurrence plots?

I am using the Webber program, but it works only to standart RP and to Cross RP...but there is nothing about joint RP. I could generate the JRP, but I don't know how to compute the RQA quantities. So, if exists a software to compute the RQA quantities with an RP input (a file with two columns (i, j), for example), it will works too.

Thank you very much.

Adriano

Re: joint recurrence plot software

Posted: Mon Nov 5, 2007 16:00
by Norbert
Can you use Matlab? Then you could use the CRP toolbox. Using this toolbox and with few lines of code you can compute RQA of JRPs. But may I ask why you need to compute the RQA measures from JRPs? What is your goal?

Norbert

Re: joint recurrence plot software

Posted: Thu Nov 8, 2007 20:32
by adrianopgpg
Indeed, Professor Marwan...It is a good question. Maybe it is a vicious behavior of the new students of RP's. We are always trying to analyze the things trough the RQA measures.
I faced the RP there are only 3 months, and my goal is to compare a financial time series with itself (but surrogated) and with a series constructed deterministically (with some criterion). But I think that a good comparison could be, beyond their separately RQA measures comparison, a cross RP or a joint RP. I don't know if there are great sense in all this, but i'm thinking yet.

Thanks for the answer an for the question!


Adriano

Universidade Estadual de Ponta Grossa - Brazil

Re: joint recurrence plot software

Posted: Fri Nov 9, 2007 10:45
by Norbert
First, I forgot to mention that you can also use the "commandline recurrence plot" tool, which allows to calculate RQA from bivariate JRPs.

However, for your purpose it could be more interesting to compare the diagonal-wise computed RR of the two data series, as it is performed to study phase-synchronisation by means of RPs. Have a look in N. Marwan, M. C. Romano, M. Thiel, J. Kurths: Recurrence Plots for the Analysis of Complex Systems, Physics Reports, 438(5-6), 237-329, DOI:10.1016/j.physrep.2006.11.001, 2007 (Section 3.8.3).

Best regards
Norbert
(I'm not a professor)

Re: joint recurrence plot software

Posted: Sun Apr 12, 2009 07:26
by Shreyas
Dear Norbert,

I am interested in JRP as my data is multivariate in nature.
However, I dint actually how to compute the JRP`s in CRP toolbox in Matlab and the Commandline RP tool.
Can you please provide suitable examples for the calculation of the JRP-RQA paramaters.
I could not get an idea so as how to compute the JRP-RQA parameters from the given literature of the CRP Toolbox and the CRP Standalone Tool.

My Inputformat consists of a file in which each column represents a time series.

Please advice me how to proceed with using the CRP toolbox and the CRP Standalone Tool.

Thanks in Advance.

Shreyas

Re: joint recurrence plot software

Posted: Wed Apr 15, 2009 15:43
by Norbert
Shreyas wrote:Dear Norbert,

I am interested in JRP as my data is multivariate in nature.
However, I dint actually how to compute the JRP`s in CRP toolbox in Matlab and the Commandline RP tool.
Can you please provide suitable examples for the calculation of the JRP-RQA paramaters.
I could not get an idea so as how to compute the JRP-RQA parameters from the given literature of the CRP Toolbox and the CRP Standalone Tool.

My Inputformat consists of a file in which each column represents a time series.

Please advice me how to proceed with using the CRP toolbox and the CRP Standalone Tool.

Thanks in Advance.

Shreyas
Hi Shreyas,

using the CRP toolbox you can create a JRP by

Code: Select all

X = jrp(x,y);
In order to perform a RQA, you have to calculate the histograms of lengths of the diagonal and vertical line lengths:

Code: Select all

[m hl] = dl(X); % histogram hl for diagonal line length
[m hv] = tt(X); % histogram hv for vertical line length
You can now calculate the measures as follows

Code: Select all

RR = mean(X(:));
DET = sum(hl(hl > 1)) / sum(hl)
LAM = sum(hv(hv > 1)) / sum(hv)
L = mean(hl(hl > 1))
TT = mean(hv(hv > 1))
I suggest to read the help text of the functions dl and tt.

The command line tool calculates directly the RQA measures. In order to calculate these measures from a JRP, just use the option -J.

Best regards
Norbert

Re: joint recurrence plot software

Posted: Thu Jan 9, 2014 10:38
by aliciarohail
You can at least use the way back machine to have a look on the VRA web page and to download the software in Version 4.9: