Corpus ID: 218502272

A simple test for causality in complex systems

  title={A simple test for causality in complex systems},
  author={Kristian Agas{\o}ster Haaga and David Diego and Jo Brendryen and Bjarte Hannisdal},
  journal={arXiv: Applications},
We provide a new solution to the long-standing problem of inferring causality from observations without modeling the unknown mechanisms. We show that the evolution of any dynamical system is related to a predictive asymmetry that quantifies causal connections from limited observations. A built-in significance criterion obviates surrogate testing and drastically improves computational efficiency. We validate our test on numerous synthetic systems exhibiting behavior commonly occurring in nature… Expand
1 Citations

Figures from this paper

Causality in Reversed Time Series: Reversed or Conserved?
Simulations of random as well as realistically motivated network coupling patterns from brain and climate show that level of coupling reversal and conservation can be well predicted by asymmetry and anormality indices introduced based on the theoretical analysis of the problem. Expand


Detecting Causality in Complex Ecosystems
A new method, based on nonlinear state space reconstruction, that can distinguish causality from correlation is introduced, and extends to nonseparable weakly connected dynamic systems (cases not covered by the current Granger causality paradigm). Expand
Spurious causalities with transfer entropy.
  • D. Smirnov
  • Physics, Medicine
  • Physical review. E, Statistical, nonlinear, and soft matter physics
  • 2013
This work highlights typical factors leading to such "spurious couplings": unobserved state variables of the driving system, low temporal resolution, and observation errors, all of which are shown to be particular cases of a general problem: imperfect observations of states of thedriving system. Expand
Testing for causality in reconstructed state spaces by an optimized mixed prediction method.
The results showed that the proposed method of optimized mixed prediction was able to reveal the presence and the direction of coupling and distinguish causality from mere correlation as well. Expand
Surrogate data for hypothesis testing of physical systems
Abstract The availability of time series of the evolution of the properties of physical systems is increasing, stimulating the development of many novel methods for the extraction of informationExpand
Comparison of six methods for the detection of causality in a bivariate time series.
Six causality detection methods were compared, which differed considerably in their capability to reveal the presence and the direction of coupling and to distinguish causality from mere correlation. Expand
Distinguishing time-delayed causal interactions using convergent cross mapping
This work expands upon the technique of CCM by explicitly considering time lags, and demonstrates the ability to identify different time-delayed interactions, distinguish between synchrony induced by strong unidirectional-forcing and true bidirectional causality, and resolve transitive causal chains. Expand
Testing for nonlinearity in time series: the method of surrogate data
A statistical approach for identifying nonlinearity in time series which is demonstrated for numerical data generated by known chaotic systems, and applied to a number of experimental time series, which arise in the measurement of superfluids, brain waves, and sunspots. Expand
Non-parametric inference of causal interactions from geological records
Quantification of drive-response relationships from geological records is challenged both by the complexity of the interactions and by indirect and uncertain proxy data. Here I demonstrate the use ofExpand
Analyzing multiple nonlinear time series with extended Granger causality
Identifying causal relations among simultaneously acquired signals is an important problem in multivariate time series analysis. For linear stochastic systems Granger proposed a simple procedureExpand
Kernel method for nonlinear granger causality.
The method performs linear Granger causality in the feature space of suitable kernel functions, assuming arbitrary degree of nonlinearity, and develops a new strategy to cope with the problem of overfitting, based on the geometry of reproducing kernel Hilbert spaces. Expand