RECURRENCE PLOTS::Recurrence Quantification Analysis, Definition

Quantification of Recurrence Plots (Recurrence Quantification Analysis)

Definition

Recurrence Quantification Analysis - The recurrence quantification analysis (RQA) is a method of nonlinear data analysis which quantifies the number and duration of recurrences of a dynamical system presented by its state space trajectory.

A quantification of recurrence plots was developed by Zbilut and Webber Jr. (Zbilut and Webber Jr., 1992; Webber Jr. and Zbilut, 1994) and extended with new measures of complexity by Marwan et al. (2002). Measures which base on diagonal structures are able to find chaos-order transitions (Trulla et al., 1996), measures based on vertical (horizontal) structures are able to find chaos-chaos transitions (laminar phases, Marwan et al., 2002).

These measures can be computed in windows along the main diagonal. This allows us to study their time dependence and can be used for the detection of transitions (Trulla et al., 1996). Another possibility is to define these measures for each diagonal parallel to the main diagonal separately (Marwan and Kurths, 2002). This approach enables the study of time delays, unstable periodic orbits (UPOs; Lathrop and Kostelich, 1989; Gilmore, 1998), and by applying to cross recurrence plots, the assessment of similarities between processes (Marwan and Kurths, 2002).

Measure

Definition

Recurrence rate RR

The percentage of recurrence points in an RP: $RR = sum_{i,j} (RP(i,j)) / N^2$ Corresponds to the correlation sum.

Determinism DET

The percentage of recurrence points which form diagonal lines: $DET = sum_{l=lmin..N} (P(l)) / sum_{l=1...N}(P(l))$ P(l) is the histogram of the lengths l of the diagonal lines.

Laminarity LAM

The percentage of recurrence points which form vertical lines: $LAM = sum_{v=vmin..N} (P(v)) / sum_{v=1...N}(P(v))$ P(v) is the histogram of the lengths v of the vertical lines.

Ratio RATIO

The ratio between DET and RR: RATIO

Averaged diagonal line length L

The average length of the diagonal lines: $L = sum_{l=lmin..N} (l*P(l)) / sum_{l=lmin...N}(P(l))$

Trapping time TT

The average length of the vertical lines: $TT = sum_{v=vmin..N} (v*P(v)) / sum_{v=vmin...N}(P(v))$

Longest diagonal line L_max

The length of the longest diagonal line LMAX

Longest vertical line V_max

The length of the longest vertical line VMAX

Divergence DIV

The inverse of L_max DIV = 1/LMAX

Related with the KS entropy of the system, i.e. with the sum of the positive Lyapunov exponents.

Entropy ENTR

The Shannon entropy of the probability distribution of the diagonal line lengths p(l): ENTR = -sum(p(l) log(p(l)))

Trend TREND

The paling of the RP towards its edges: $TREND = (sum_{i=1...N}(i-N/2)(RR_i - <RR_i>)) / (sum_{i=1...N}(i-N/2)^2)$

N – number of points on the phase space trajectory

N_l – number of diagonal lines in the recurrence plot

N_v – number of vertical lines in the recurrence plot

P(l), P(v) – histogram of the line lengths of diagonal/ vertical lines

Ñ – maximal number of diagonals parallel to the LOI which will be considered for the calculation of TREND

» Recurrence quantification analysis in SciTopics
» Recurrence quantification analysis in Wikipedia
» Further definitions of recurrence quantification analysis (Google)

Please respect the copyrights! The content of this web site is protected by a Creative Commons License. You may use the text or figures, but you have to cite this source (www.recurrence-plot.tk) as well as N. Marwan: Encounters With Neighbours - Current Developments Of Concepts Based On Recurrence Plots And Their Applications, Ph.D. Thesis, University of Potsdam, ISBN 3-00-012347-4.

RECURRENCE PLOTS AND CROSS RECURRENCE PLOTS

Quantification of Recurrence Plots (Recurrence Quantification Analysis)

Definition