Method References
This page is generated from MethodRegistry.list_catalog() so citations,
upstream package links, and method metadata stay aligned.
Current package version target: 0.1.1.
These links cover the method families and upstream packages used or compared
in the public De-Time workflow surface. MA_BASELINE is an in-package smoke
baseline and therefore has no separate upstream citation.
Flagship methods
MSSA
- Summary: Multivariate SSA for shared-structure decomposition across channels.
- Optional/runtime dependencies: none
Primary references: - Golyandina and Zhigljavsky (2020), Singular Spectrum Analysis for Time Series - Primary SSA/MSSA reference used for the multivariate extension.
Related packages: - SSALib - SSA-focused package; useful comparison point for SSA-family workflows.
SSA
- Summary: Singular spectrum analysis for structured univariate decomposition.
- Optional/runtime dependencies: none
Primary references: - Golyandina and Zhigljavsky (2020), Singular Spectrum Analysis for Time Series - Primary SSA reference; the second edition also covers multivariate SSA (MSSA).
Related packages: - SSALib - Specialist SSA package used as an external comparison point.
STD
- Summary: Fast seasonal-trend decomposition with dispersion-aware diagnostics.
- Optional/runtime dependencies: none
Primary references: - Dudek (2022), STD: A Seasonal-Trend-Dispersion Decomposition of Time Series - Primary reference for STD and the robust seasonal-trend-dispersion family.
Related packages: - none declared
STDR
- Summary: Robust seasonal-trend decomposition for noisier periodic signals.
- Optional/runtime dependencies: none
Primary references: - Dudek (2022), STD: A Seasonal-Trend-Dispersion Decomposition of Time Series - Primary reference for STD and the robust seasonal-trend-dispersion family.
Related packages: - none declared
Stable wrappers and retained methods
CEEMDAN
- Summary: Noise-assisted EMD variant for more stable IMF extraction.
- Optional/runtime dependencies: PyEMD
Primary references: - Torres et al. (2011), A complete ensemble empirical mode decomposition with adaptive noise - PyEMD CEEMDAN docs cite the original ICASSP 2011 paper. - Colominas, Schlotthauer, and Torres (2014), Improved complete ensemble EMD: A suitable tool for biomedical signal processing - Improved CEEMDAN variant adopted by the PyEMD implementation used by De-Time.
Related packages: - PyEMD - Upstream Python package wrapped by De-Time for EMD-family methods.
EMD
- Summary: Empirical mode decomposition under the De-Time result contract.
- Optional/runtime dependencies: PyEMD
Primary references: - Huang et al. (1998), The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis - Primary empirical mode decomposition reference.
Related packages: - PyEMD - Upstream Python package wrapped by De-Time for EMD-family methods.
MA_BASELINE
- Summary: Simple moving-average baseline for smoke tests and lightweight workflows.
- Optional/runtime dependencies: none
Primary references: - none declared
Related packages: - none declared
MSTL
- Summary: Statsmodels MSTL wrapped into the De-Time workflow surface.
- Optional/runtime dependencies: statsmodels
Primary references: - Bandara, Hyndman, and Bergmeir (2021), MSTL: A Seasonal-Trend Decomposition Algorithm for Time Series with Multiple Seasonal Patterns - Primary MSTL reference used by the statsmodels implementation.
Related packages: - statsmodels - Official project site for the upstream MSTL implementation.
ROBUST_STL
- Summary: Robust STL-style decomposition wrapped for reproducible workflows.
- Optional/runtime dependencies: statsmodels
Primary references: - Cleveland et al. (1990), STL: A Seasonal-Trend Decomposition Procedure Based on LOESS - Robust STL in De-Time builds on the same STL literature and upstream implementation family.
Related packages: - statsmodels - Official project site for the upstream STL implementation family.
STL
- Summary: Classical STL wrapped into the De-Time workflow contract.
- Optional/runtime dependencies: statsmodels
Primary references: - Cleveland et al. (1990), STL: A Seasonal-Trend Decomposition Procedure Based on LOESS - Statsmodels STL docs cite the original Journal of Official Statistics paper.
Related packages: - statsmodels - Official project site for the upstream STL implementation.
VMD
- Summary: Variational mode decomposition integrated into the common workflow layer.
- Optional/runtime dependencies: vmdpy, sktime
Primary references: - Dragomiretskiy and Zosso (2014), Variational Mode Decomposition - Primary variational mode decomposition reference.
Related packages:
- sktime - Current maintained ecosystem for vmdpy, which the archived project directs users toward.
- vmdpy - Archived Python VMD package used by the current De-Time wrapper.
WAVELET
- Summary: Wavelet-based decomposition exposed through the common output contract.
- Optional/runtime dependencies: PyWavelets
Primary references: - Mallat (1989), A theory for multiresolution signal decomposition: the wavelet representation - Foundational wavelet multiresolution reference. - Lee et al. (2019), PyWavelets: A Python package for wavelet analysis - Package citation for the upstream wavelet implementation used by De-Time.
Related packages: - PyWavelets - Official documentation for the upstream wavelet package.
Optional backend methods
MEMD
- Summary: Optional multivariate EMD backend for shared oscillatory structure.
- Optional/runtime dependencies: PySDKit
Primary references: - Rehman and Mandic (2010), Multivariate empirical mode decomposition - Primary MEMD reference for the multivariate EMD extension.
Related packages: - PySDKit - Optional multivariate backend used by De-Time for MEMD.
MVMD
- Summary: Optional multivariate VMD backend for shared frequency structure.
- Optional/runtime dependencies: PySDKit
Primary references: - Rehman and Aftab (2019), Multivariate Variational Mode Decomposition - Primary MVMD reference for the multivariate VMD extension.
Related packages: - PySDKit - Optional multivariate backend used by De-Time for MVMD.
Experimental methods
GABOR_CLUSTER
- Summary: Experimental clustering-based decomposition path.
- Optional/runtime dependencies: faiss
Primary references: - Gabor (1946), Theory of Communication - Historical reference for the Gabor time-frequency representation family. - Douze et al. (2024), The Faiss library - Reference for the similarity-search backend used by the experimental clustering path.
Related packages: - Faiss - Vector similarity search library required by the experimental clustering backend.