Time Series

41 public types in this namespace, organized by kind.

Models & Types (34)

Type Summary
ARIMAModel<T> Implements an ARIMA (AutoRegressive Integrated Moving Average) model for time series forecasting.
ARIMAXModel<T> Implements an ARIMAX (AutoRegressive Integrated Moving Average with eXogenous variables) model for time series forecasting.
ARMAModel<T> Implements an ARMA (AutoRegressive Moving Average) model for time series forecasting.
ARModel<T> Implements an AR (AutoRegressive) model for time series forecasting.
AutoformerModel<T> Implements the Autoformer model for long-term time series forecasting with decomposition.
BayesianStructuralTimeSeriesModel<T> Implements a Bayesian Structural Time Series model for flexible time series forecasting.
ChronosFoundationModel<T> Implements Chronos-T5 (Ansari et al.
DLinearModel<T> DLinear — decomposition-linear forecaster (Zeng et al., AAAI 2023, “Are Transformers Effective for Time Series Forecasting?”).
DeepANT<T> Implements DeepAnT (Munir et al.
DeepARModel<T> Implements DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks.
DynamicRegressionWithARIMAErrors<T> Implements a Dynamic Regression model with ARIMA errors for time series forecasting.
ExponentialSmoothingModel<T> Represents a model that implements exponential smoothing for time series forecasting.
GARCHModel<T> Represents a Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model for time series with changing volatility.
InformerModel<T> Implements the Informer model for efficient long-sequence time series forecasting.
InterventionAnalysisModel<T> Represents a model that analyzes and forecasts time series data with interventions or structural changes.
LSTMVAE<T> Implements LSTM-VAE (Long Short-Term Memory Variational Autoencoder) for anomaly detection.
MAModel<T> Implements a Moving Average (MA) model for time series forecasting.
NBEATSModel<T> Implements the N-BEATS (Neural Basis Expansion Analysis for Time Series) model for forecasting.
NHiTSModel<T> Implements N-HiTS (Neural Hierarchical Interpolation for Time Series) for efficient long-horizon forecasting.
NLinearModel<T> NLinear — normalization-linear forecaster (Zeng et al., AAAI 2023).
NeuralNetworkARIMAModel<T> Represents a Neural Network ARIMA (Autoregressive Integrated Moving Average) model for time series forecasting.
ProphetModel<T, TInput, TOutput> Represents a Prophet model for time series forecasting.
SARIMAModel<T> Implements a Seasonal Autoregressive Integrated Moving Average (SARIMA) model for time series forecasting.
STLDecomposition<T> Implements Seasonal-Trend decomposition using LOESS (STL) for time series analysis.
SpectralAnalysisModel<T> Implements spectral analysis for time series data, which transforms time domain signals into the frequency domain.
StateSpaceModel<T> Implements a State Space Model for time series analysis and forecasting.
TBATSModel<T> Implements the TBATS (Trigonometric, Box-Cox transform, ARMA errors, Trend, and Seasonal components) model for complex time series forecasting with multiple seasonal patterns.
TemporalFusionTransformer<T> Implements the Temporal Fusion Transformer (TFT) per Lim et al.
TiDEModel<T> TiDE — Time-series Dense Encoder (Das et al., TMLR 2023).
TimeSeriesIsolationForest<T> Implements Isolation Forest for time series anomaly detection.
TransferFunctionModel<T> Implements a Transfer Function Model for time series analysis, which combines ARIMA modeling with external input variables to capture dynamic relationships between multiple time series.
UnobservedComponentsModel<T, TInput, TOutput> Implements an Unobserved Components Model (UCM) for time series decomposition and forecasting.
VARMAModel<T> Implements a Vector Autoregressive Moving Average (VARMA) model for multivariate time series forecasting.
VectorAutoRegressionModel<T> Implements a Vector Autoregression (VAR) model for multivariate time series forecasting.

Base Classes (1)

Type Summary
TimeSeriesModelBase<T> Provides a base class for all time series forecasting models in the library.

Enums (2)

Type Summary
DeepARLikelihood The likelihood of a DeepAR network’s observations (Salinas et al.
InformerAttention The attention of Informer’s encoder and decoder self-attention (Zhou et al.

Options & Configuration (4)

Type Summary
ChronosOptions<T> Options for ChronosFoundationModel.
DeepANTOptions<T> Options for the DeepAnT model.
DeepAntConfiguration The sizes of DeepAnT’s time-series predictor.
LSTMVAEOptions<T> Options for LSTM-VAE model.