Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Complex Systems›Recurrence Quantification Analysis (RQA)
Machine learningNonlinear dynamics

Recurrence Quantification Analysis (RQA)

Also known as: RQA, Recurrence Plot Analysis, Nonlinear Recurrence Analysis, Tekrarlama Kantifikasyon Analizi

Recurrence Quantification Analysis (RQA) is a nonlinear method for characterizing the dynamics of a time series by quantifying the small-scale structure of its recurrence plot. Introduced in its modern, comprehensive form by Marwan, Romano, Thiel, and Kurths in 2007, RQA extracts scalar measures — such as recurrence rate, determinism, laminarity, and Shannon entropy — that capture periodicity, chaos, stationarity, and transitions in complex dynamical systems.

ScholarGate
  1. Machine learning
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Recurrence Quantification Analysis
Fractal AnalysisSample EntropyConvergent Cross MappingSelf-Organized Criticali…

When to use it

RQA is appropriate when the time series is suspected to be nonlinear, non-stationary, or of unknown dynamics — contexts where classical spectral methods or linear autocorrelation are insufficient. It suits relatively short records (hundreds to low thousands of points) from physiological, geophysical, financial, or engineering sources. Key assumptions are that the data arise from a deterministic or stochastic dynamical system and that embedding parameters (m, tau, epsilon) can be estimated from the data. Limitations include sensitivity to the threshold epsilon, computational cost scaling as O(N^2), and interpretive ambiguity when signal length is short. Alternatives include sample entropy, detrended fluctuation analysis, or wavelet-based nonlinear methods.

Strengths & limitations

Strengths
  • Works with short, noisy, non-stationary time series where Fourier or linear AR methods fail.
  • Produces multiple interpretable scalar indices (RR, DET, LAM, ENTR, TT) from a single analysis.
  • No assumption of stationarity or linearity; applicable to real-world complex systems.
  • Recurrence plots provide an intuitive visual complement to the quantitative measures.
Limitations
  • Results are sensitive to the choice of embedding dimension m, time delay tau, and threshold epsilon — poor choices distort all derived measures.
  • Computational complexity is O(N^2) in memory and time, limiting applicability to very long time series without windowing strategies.
  • No universal threshold-selection criterion exists; practice often relies on heuristics (e.g., 10% recurrence rate) that may not generalize.
  • Statistical significance testing of RQA measures requires careful surrogate-data procedures, adding analytical overhead.

Frequently asked

How do I choose the threshold epsilon for the recurrence matrix?

A common heuristic is to select epsilon such that the recurrence rate (RR) falls between 2% and 10% of all possible point pairs, ensuring the plot is neither too sparse nor too dense. Alternatively, epsilon can be set as a fixed percentage of the attractor diameter or chosen via a fixed nearest-neighbor fraction. Sensitivity analysis across a range of epsilon values is strongly recommended before interpreting results.

What is the minimum time series length required for reliable RQA?

There is no strict universal minimum, but most practitioners recommend at least N = 500 data points after embedding to ensure sufficient diagonal and vertical line statistics. For very short series (N < 200), RQA measures exhibit high variance and should be interpreted cautiously, ideally supplemented by bootstrap confidence intervals or surrogate testing.

How does RQA differ from sample entropy?

Both methods quantify regularity in time series without assuming linearity, but they differ in structure. Sample entropy counts template matches up to a fixed tolerance and length, producing a single scalar reflecting predictability. RQA constructs a full recurrence matrix and extracts multiple measures (RR, DET, LAM, ENTR, TT), each capturing a distinct aspect of the dynamics — determinism, laminarity, and entropy of line-length distributions — providing a richer, multi-faceted characterization.

Sources

  1. Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics Reports, 438(5–6), 237–329. DOI: 10.1016/j.physrep.2006.11.001 ↗

How to cite this page

ScholarGate. (2026, June 2). Recurrence Quantification Analysis (RQA). ScholarGate. https://scholargate.app/en/complex-systems/recurrence-quantification-analysis

Related methods

Fractal AnalysisSample Entropy

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Fractal AnalysisComplex Systems↔ compare
  • Sample EntropyComplex Systems↔ compare
Compare side by side →

Referenced by

Convergent Cross MappingFractal AnalysisSample EntropySelf-Organized Criticality

Similar methods

Sample EntropyFractal AnalysisConvergent Cross MappingSingular Spectrum AnalysisTransfer EntropyNonlinear Granger CausalityHilbert-Huang TransformThreshold and Smooth-Transition VAR

Related reference concepts

AttractorsChaos TheoryDynamical SystemsErgodic TheoryReceiver Operating Characteristic CurveBifurcation Theory

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Recurrence Quantification Analysis (Recurrence Quantification Analysis (RQA)). Retrieved 2026-07-21 from https://scholargate.app/en/complex-systems/recurrence-quantification-analysis · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Marwan, Romano, Thiel & Kurths
Year
2007
Type
Nonlinear time-series characterization
Subfamily
Nonlinear dynamics
Data Requirement
Univariate or multivariate time series
Output
Scalar recurrence statistics (RR, DET, LAM, ENTR, etc.)
Related methods
Fractal AnalysisSample Entropy
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account