<br />

|---|
| [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png) View source on GitHub](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/sts/__init__.py) |

Framework for Bayesian structural time series models.

See the [blog post](https://blog.tensorflow.org/2019/03/structural-time-series-modeling-in.html)
for an introductory example.

## Classes

[`class AdditiveStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/AdditiveStateSpaceModel): A state space model representing a sum of component state space models.

[`class Autoregressive`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/Autoregressive): Formal representation of an autoregressive model.

[`class AutoregressiveIntegratedMovingAverage`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/AutoregressiveIntegratedMovingAverage): Represents an autoregressive integrated moving-average (ARIMA) model.

[`class AutoregressiveMovingAverageStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/AutoregressiveMovingAverageStateSpaceModel): State space model for an autoregressive moving average process.

[`class AutoregressiveStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/AutoregressiveStateSpaceModel): State space model for an autoregressive process.

[`class ConstrainedSeasonalStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/ConstrainedSeasonalStateSpaceModel): Seasonal state space model with effects constrained to sum to zero.

[`class DynamicLinearRegression`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/DynamicLinearRegression): Formal representation of a dynamic linear regresson model.

[`class DynamicLinearRegressionStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/DynamicLinearRegressionStateSpaceModel): State space model for a dynamic linear regression from provided covariates.

[`class IntegratedStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/IntegratedStateSpaceModel): Integrates (/cumsums) a noise-free state space model.

[`class LinearRegression`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/LinearRegression): Formal representation of a linear regression from provided covariates.

[`class LocalLevel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/LocalLevel): Formal representation of a local level model.

[`class LocalLevelStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/LocalLevelStateSpaceModel): State space model for a local level.

[`class LocalLinearTrend`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/LocalLinearTrend): Formal representation of a local linear trend model.

[`class LocalLinearTrendStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/LocalLinearTrendStateSpaceModel): State space model for a local linear trend.

[`class MaskedTimeSeries`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/MaskedTimeSeries): Named tuple encoding a time series `Tensor` and optional missingness mask.

[`class MissingValuesTolerance`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/MissingValuesTolerance): MissingValuesTolerance(overall_fraction, fraction_low_missing_number, fraction_high_missing_number, low_missing_number, high_missing_number)

[`class Seasonal`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/Seasonal): Formal representation of a seasonal effect model.

[`class SeasonalStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SeasonalStateSpaceModel): State space model for a seasonal effect.

[`class SemiLocalLinearTrend`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SemiLocalLinearTrend): Formal representation of a semi-local linear trend model.

[`class SemiLocalLinearTrendStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SemiLocalLinearTrendStateSpaceModel): State space model for a semi-local linear trend.

[`class SmoothSeasonal`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SmoothSeasonal): Formal representation of a smooth seasonal effect model.

[`class SmoothSeasonalStateSpaceModel`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SmoothSeasonalStateSpaceModel): State space model for a smooth seasonal effect.

[`class SparseLinearRegression`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/SparseLinearRegression): Formal representation of a sparse linear regression.

[`class StructuralTimeSeries`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/StructuralTimeSeries): Base class for structural time series models.

[`class Sum`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/Sum): Sum of structural time series components.

## Functions

[`build_factored_surrogate_posterior(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/build_factored_surrogate_posterior): Build a variational posterior that factors over model parameters.

[`build_factored_surrogate_posterior_stateless(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/build_factored_surrogate_posterior_stateless): Returns stateless functions for building a variational posterior.

[`decompose_by_component(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/decompose_by_component): Decompose an observed time series into contributions from each component.

[`decompose_forecast_by_component(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/decompose_forecast_by_component): Decompose a forecast distribution into contributions from each component.

[`fit_with_hmc(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/fit_with_hmc): Draw posterior samples using Hamiltonian Monte Carlo (HMC).

[`forecast(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/forecast): Construct predictive distribution over future observations.

[`impute_missing_values(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/impute_missing_values): Runs posterior inference to impute the missing values in a time series. (deprecated argument values)

[`moments_of_masked_time_series(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/moments_of_masked_time_series): Compute mean and variance, accounting for a mask.

[`one_step_predictive(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/one_step_predictive): Compute one-step-ahead predictive distributions for all timesteps. (deprecated argument values)

[`regularize_series(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/regularize_series): Infers frequency and makes an irregular time series regular.

[`sample_uniform_initial_state(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/sts/sample_uniform_initial_state): Initialize from a uniform \[-2, 2\] distribution in unconstrained space. (deprecated)