Reference library
Surrogate time series
Paper summary
Surrogate tests compare a declared null hypothesis with data, rather than establish chaos whenever traces differ. Nonlinear observation of a linear stochastic process can create bursts. AAFT can bias spectra; iterative refinement improves constraint matching but requires residual checks. Nonstationarity and external kicks can cause null rejection. Nonlinearity testing does not establish determinism, a finite attractor dimension or chaos.
Relevance to the project
Assistant interpretation: protects analysis of adaptive neural traces against labeling drift, correlated noise, bursts or argmax effects as chaos. Nulls should preserve the input/measurement constraints relevant to the question; known transition access offers stronger interventions than scalar surrogate testing alone.
Selected methods and interpretation reading, not complete paper/appendix implementation audit. Statistical conclusions depend on null choice, stationarity, data length and test selection; a failed rejection is not proof of linearity. Artifact compilation date retained explicitly without treating this as new 2024 research.
Source record
- Reference ID
- surrogate-time-series-2000
- Archived version
- Served arXiv:chao-dyn/9909037v1 artifact; identifier27Sep1999, displayed compilation date26Nov2024; associated2000 article, not final journal typesetting
- Summary provenance
- reference-library/papers/surrogate-time-series-2000/metadata.json · summary
- Review status
- source-scoped reading; not replication
- Reading scope
- PDF pages 2–4,7–8,20,23–25 read; page 1 visually checked for identity/version. Other sections not fully read; TISEAN code not inspected or run.