tfp.sts.StructuralTimeSeries

Class StructuralTimeSeries

Base class for structural time series models.

A StructuralTimeSeries object represents a declarative specification of a structural time series model, including priors on model parameters. It implements a joint probability model p(params, y) = p(params) p(y | params), where params denotes a list of real-valued parameters specified by the child class, and p(y | params) is a linear Gaussian state space model with structure determined by the child class.

__init__

__init__(
    parameters,
    latent_size,
    name='StructuralTimeSeries'
)

Construct a specification for a structural time series model.

Args:

  • parameters: list of Parameter namedtuples, each specifying the name and prior distribution of a model parameter along with a bijective transformation from an unconstrained space to the support of that parameter. The order of this list determines the canonical parameter ordering used by fitting and inference algorithms.
  • latent_size: Python int specifying the dimensionality of the latent state space for this model.
  • name: Python str name for this model component.

Properties

batch_shape

Static batch shape of models represented by this component.

Returns:

  • batch_shape: A tf.TensorShape giving the broadcast batch shape of all model parameters. This should match the batch shape of derived state space models, i.e., self.make_state_space_model(...).batch_shape. It may be partially defined or unknown.

latent_size

Python int dimensionality of the latent space in this model.

name

Name of this model component.

parameters

List of Parameter(name, prior, bijector) namedtuples for this model.

Methods

batch_shape_tensor

batch_shape_tensor()

Runtime batch shape of models represented by this component.

Returns:

  • batch_shape: int Tensor giving the broadcast batch shape of all model parameters. This should match the batch shape of derived state space models, i.e., self.make_state_space_model(...).batch_shape_tensor().

joint_log_prob

joint_log_prob(observed_time_series)

Build the joint density log p(params) + log p(y|params) as a callable.

Args:

  • observed_time_series: Observed Tensor trajectories of shape sample_shape + batch_shape + [num_timesteps, 1] (the trailing 1 dimension is optional if num_timesteps > 1), where batch_shape should match self.batch_shape (the broadcast batch shape of all priors on parameters for this structural time series model).

Returns:

log_joint_fn: A function taking a Tensor argument for each model parameter, in canonical order, and returning a Tensor log probability of shape batch_shape. Note that, unlike tfp.Distributions log_prob methods, the log_joint sums over the sample_shape from y, so that sample_shape does not appear in the output log_prob. This corresponds to viewing multiple samples in y as iid observations from a single model, which is typically the desired behavior for parameter inference.

make_state_space_model

make_state_space_model(
    num_timesteps,
    param_vals=None,
    initial_state_prior=None,
    initial_step=0
)

Instantiate this model as a Distribution over specified num_timesteps.

Args:

  • num_timesteps: Python int number of timesteps to model.
  • param_vals: a list of Tensor parameter values in order corresponding to self.parameters, or a dict mapping from parameter names to values.
  • initial_state_prior: an optional Distribution instance overriding the default prior on the model's initial state. This is used in forecasting ("today's prior is yesterday's posterior").
  • initial_step: optional int specifying the initial timestep to model. This is relevant when the model contains time-varying components, e.g., holidays or seasonality.

Returns:

  • dist: a LinearGaussianStateSpaceModel Distribution object.

prior_sample

prior_sample(
    num_timesteps,
    initial_step=0,
    params_sample_shape=(),
    trajectories_sample_shape=(),
    seed=None
)

Sample from the joint prior over model parameters and trajectories.

Args:

  • num_timesteps: Scalar int Tensor number of timesteps to model.
  • initial_step: Optional scalar int Tensor specifying the starting timestep. Default value: 0.
  • params_sample_shape: Number of possible worlds to sample iid from the parameter prior, or more generally, Tensor int shape to fill with iid samples. Default value: .
  • trajectories_sample_shape: For each sampled set of parameters, number of trajectories to sample, or more generally, Tensor int shape to fill with iid samples. Default value: .
  • seed: Python int random seed.

Returns:

  • trajectories: float Tensor of shape trajectories_sample_shape + params_sample_shape + [num_timesteps, 1] containing all sampled trajectories.
  • param_samples: list of sampled parameter value Tensors, in order corresponding to self.parameters, each of shape params_sample_shape + prior.batch_shape + prior.event_shape.