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 DeTime 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 DeTime.

Related packages: - PyEMD - Upstream Python package wrapped by DeTime for EMD-family methods.

EMD

  • Summary: Empirical mode decomposition under the DeTime 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 DeTime 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 DeTime 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 DeTime 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 DeTime 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 DeTime 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 DeTime.

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 DeTime 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 DeTime for MVMD.

Experimental methods

AMD_BLOCK

  • Summary: AMD-inspired multiscale smoothing head with periodic-template seasonal reconstruction.
  • Optional/runtime dependencies: none

Primary references: - Hu et al. (2024), Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting - Source framework for adaptive multiscale decomposition.

Related packages: - none declared

AUTOFORMER_BLOCK

  • Summary: Standalone moving-average decomposition head extracted from the Autoformer architecture.
  • Optional/runtime dependencies: none

Primary references: - Wu et al. (2021), Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting - Source architecture for the moving-average decomposition block exposed by AUTOFORMER_BLOCK.

Related packages: - none declared

DELELSTM_BLOCK

  • Summary: DeLELSTM-inspired Holt trend plus periodic-template seasonal decomposition head.
  • Optional/runtime dependencies: none

Primary references: - Wang et al. (2023), DeLELSTM: Decomposition-based Linear Explainable LSTM to Capture Instantaneous and Long-term Effects in Time Series - Source model for decomposition-based explainable LSTM effects.

Related packages: - none declared

DLINEAR_BLOCK

  • Summary: Standalone moving-average decomposition head extracted from DLinear-style forecasting blocks.
  • Optional/runtime dependencies: none

Primary references: - Zeng et al. (2023), Are Transformers Effective for Time Series Forecasting? - Introduces the LTSF-Linear family, including the DLinear decomposition-based linear model.

Related packages: - none declared

FREQMOE_BLOCK

  • Summary: FreqMoE-inspired frequency mixture head for trend plus multi-band seasonal reconstruction.
  • Optional/runtime dependencies: none

Primary references: - Liu (2025), FreqMoE: Enhancing Time Series Forecasting through Frequency Decomposition Mixture of Experts - Source architecture for frequency decomposition mixture-of-experts forecasting.

Related packages: - none declared

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.

INPARFORMER_BLOCK

  • Summary: InParformer-inspired moving-average trend plus periodic-template seasonal decomposition head.
  • Optional/runtime dependencies: none

Primary references: - Cao et al. (2023), InParformer: Evolutionary Decomposition Transformers with Interactive Parallel Attention for Long-Term Time Series Forecasting - Source architecture for evolutionary seasonal-trend decomposition in a transformer forecaster.

Related packages: - none declared

LEDDAM_BLOCK

  • Summary: LEDDAM LD smoothing block exposed as a Gaussian-kernel decomposition operator.
  • Optional/runtime dependencies: none

Primary references: - Yu et al. (2024), Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling - Introduces LEDDAM, the learnable decomposition and dual-attention module.

Related packages: - none declared

MOVING_AVERAGE_DECOMPOSITION_BLOCK

  • Summary: Generic neural forecasting moving-average decomposition block exposed as a DeTime method.
  • Optional/runtime dependencies: none

Primary references: - Wu et al. (2021), Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting - Primary source for treating moving-average series decomposition as an internal neural forecasting block. - Zeng et al. (2023), Are Transformers Effective for Time Series Forecasting? - Uses decomposition-based linear forecasting as a simple long-term forecasting baseline.

Related packages: - none declared

NBEATS_INTERPRETABLE

  • Summary: Torch-backed interpretable N-BEATS trend and seasonality stacks used as a learned decomposition prior.
  • Optional/runtime dependencies: torch

Primary references: - Oreshkin et al. (2020), N-BEATS: Neural basis expansion analysis for interpretable time series forecasting - Source for interpretable trend and seasonality basis stacks.

Related packages: - none declared

PARSIMONY_BLOCK

  • Summary: Parsimony-inspired trend head with compact harmonic seasonal projection.
  • Optional/runtime dependencies: none

Primary references: - Deng et al. (2024), Parsimony or Capability? Decomposition Delivers Both in Long-term Time Series Forecasting - Source paper for parameter-efficient decomposition in long-term forecasting.

Related packages: - none declared

ST_MTM_BLOCK

  • Summary: ST-MTM-inspired smoothing head combining trend smoothing and smoothed periodic seasonality.
  • Optional/runtime dependencies: none

Primary references: - Seo and Lim (2025), ST-MTM: Masked Time Series Modeling with Seasonal-Trend Decomposition for Time Series Forecasting - Source method for seasonal-trend masked time-series modeling.

Related packages: - none declared

TIMEKAN_BLOCK

  • Summary: TimeKAN-inspired decomposition head blending template and harmonic seasonal estimates.
  • Optional/runtime dependencies: none

Primary references: - Huang et al. (2025), TimeKAN: KAN-based Frequency Decomposition Learning Architecture for Long-term Time Series Forecasting - Source method for KAN-based frequency decomposition learning.

Related packages: - none declared

TIMES2D_BLOCK

  • Summary: Times2D-inspired multi-period harmonic decomposition head.
  • Optional/runtime dependencies: none

Primary references: - Nematirad, Pahwa, and Natarajan (2025), Times2D: Multi-Period Decomposition and Derivative Mapping for General Time Series Forecasting - Source method for multi-period decomposition and 2D time-series mapping.

Related packages: - none declared

WAVEFORM_BLOCK

  • Summary: WaveForM-inspired wavelet multiresolution decomposition head.
  • Optional/runtime dependencies: PyWavelets

Primary references: - Yang et al. (2023), WaveForM: Graph Enhanced Wavelet Learning for Long Sequence Forecasting of Multivariate Time Series - Source architecture for graph-enhanced wavelet learning.

Related packages: - none declared

WAVELETMIXER_BLOCK

  • Summary: WaveletMixer-inspired multiresolution decomposition head using mixed wavelet detail levels.
  • Optional/runtime dependencies: PyWavelets

Primary references: - Zhang et al. (2025), WaveletMixer: A Multi-Resolution Wavelets Based MLP-Mixer for Multivariate Long-Term Time Series Forecasting - Source method for multi-resolution wavelet mixer forecasting.

Related packages: - none declared

XPATCH_BLOCK

  • Summary: xPatch-inspired exponential smoothing head for standalone trend and local-season decomposition.
  • Optional/runtime dependencies: none

Primary references: - Stitsyuk and Choi (2024), xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition - Source architecture for exponential seasonal-trend decomposition.

Related packages: - none declared