canari.component.baseline_component

This module defines three components for modelling baselines including LocalLevel, LocalTrend, and LocalAcceleration.

class canari.component.baseline_component.LocalAcceleration(std_error: float | None = 0.0, mu_states: list[float] | None = None, var_states: list[float] | None = None)

Bases: BaseComponent

LocalAcceleration class, inheriting from Canari’s BaseComponent. It models baselines with a locally constant acceleration (linear speed and curvature level) over time. It has three hidden states level, trend, and acceleration.

Parameters:

• std_error (Optional[float]) – Standard deviation of the process noise. Defaults: initialized to 0.

• mu_states (Optional[list[float]]) – Initial mean of the hidden state. Defaults: initialized to zeros.

• var_states (Optional[list[float]]) – Initial variance of the hidden state. Defaults: initialized to zeros.

Examples

>>> from canari.component import LocalAcceleration
>>> # With known mu_states and var_states
>>> local_acc = LocalAcceleration(mu_states = [1, 2, 3], var_states = [1, 2, 3], std_error=0.3)
>>> # With default mu_states and var_states
>>> local_acc = LocalAcceleration(std_error=0.3)
>>> local_acc.transition_matrix
array([[1. , 1. , 0.5],
       [0. , 1. , 1. ],
       [0. , 0. , 1. ]])
>>> local_acc.observation_matrix
array([[1, 0, 0]])
>>> local_acc.process_noise_matrix
>>> local_acc.mu_states
>>> local_acc.var_states

initialize_component_name()

Initialize the component’s name. str.

initialize_mu_states()

Initialize the mean of the hidden states. Output an 2D array.

initialize_num_states()

Initialize the number of hidden states. int.

initialize_observation_matrix()

Initialize the observation matrix. Output an 2D array.

initialize_process_noise_matrix()

Initialize the process noise covariance matrix. Output an 2D array.

initialize_states_name()

Initialize the names of all hidden states. list[str].

initialize_transition_matrix()

Initialize the transition matrix. Output an 2D array.

initialize_var_states()

Initialize the covariance matrix of the hidden states. Output an 2D array.

class canari.component.baseline_component.LocalLevel(std_error: float | None = 0.0, mu_states: list[float] | None = None, var_states: list[float] | None = None)

Bases: BaseComponent

LocalLevel class, inheriting from Canari’s BaseComponent. It models baselines with a locally constant level (zero speed) over time. It has one hidden state level.

Parameters:

• std_error (Optional[float]) – Standard deviation of the process noise. Defaults: initialized to zeros 0.

• mu_states (Optional[list[float]]) – Initial mean of the hidden states. Defaults: initialized to zeros.

• var_states (Optional[list[float]]) – Initial variance of the hidden states. Defaults: initialized to zeros.

Examples

>>> from canari.component import LocalLevel
>>> # With known mu_states and var_states
>>> level = LocalLevel(mu_states=[1], var_states=[1],std_error=0.5)
>>> # With default mu_states and var_states
>>> level = LocalLevel(std_error=0.5)
>>> level.component_name
'local level'
>>> level.states_name
['level']
>>> level.transition_matrix
array([[1]])
>>> level.observation_matrix
array([[1]])
>>> level.process_noise_matrix
>>> level.mu_states
>>> level.var_states

initialize_component_name()

Initialize the component’s name. str.

initialize_mu_states()

Initialize the mean of the hidden states. Output an 2D array.

initialize_num_states()

Initialize the number of hidden states. int.

initialize_observation_matrix()

Initialize the observation matrix. Output an 2D array.

initialize_process_noise_matrix()

Initialize the process noise covariance matrix. Output an 2D array.

initialize_states_name()

Initialize the names of all hidden states. list[str].

initialize_transition_matrix()

Initialize the transition matrix. Output an 2D array.

initialize_var_states()

Initialize the covariance matrix of the hidden states. Output an 2D array.

class canari.component.baseline_component.LocalTrend(std_error: float | None = 0.0, mu_states: list[float] | None = None, var_states: list[float] | None = None)

Bases: BaseComponent

LocalTrend class, inheriting from Canari’s BaseComponent. It models baselines with a locally constant speed over time (linear level). It has two hidden states level and trend.

Parameters:

• std_error (Optional[float]) – Standard deviation of the process noise. Defaults: initialized to 0.

• mu_states (Optional[list[float]]) – Initial mean of the hidden state. Defaults: initialized to zeros.

• var_states (Optional[list[float]]) – Initial variance of the hidden state. Defaults: initialized to zeros.

Examples

>>> from canari.component import LocalTrend
>>> # With known mu_states and var_states
>>> local_trend = LocalAcceleration(mu_states=[1, 2], var_states=[1, 2], std_error=0.3)
>>> # With default mu_states and var_states
>>> local_trend = LocalTrend(std_error=0.2)
>>> level.component_name
'local trend'
>>> local_trend.states_name
['level', 'trend']
>>> local_trend.transition_matrix
array([[1, 1],
       [0, 1]])
>>> local_trend.observation_matrix
array([[1, 0]])
>>> local_trend.process_noise_matrix
>>> local_trend.mu_states
>>> local_trend.var_states

initialize_component_name()

Initialize the component’s name. str.

initialize_mu_states()

Initialize the mean of the hidden states. Output an 2D array.

initialize_num_states()

Initialize the number of hidden states. int.

initialize_observation_matrix()

Initialize the observation matrix. Output an 2D array.

initialize_process_noise_matrix()

Initialize the process noise covariance matrix. Output an 2D array.

initialize_states_name()

Initialize the names of all hidden states. list[str].

initialize_transition_matrix()

Initialize the transition matrix. Output an 2D array.

initialize_var_states()

Initialize the covariance matrix of the hidden states. Output an 2D array.