chronax.models.theta.auto

`AutoTheta`

chronax.models.AutoTheta · inherits BaseForecaster

AutoTheta model. Automatically selects the best Theta model variant (STM, OTM, DSTM, DOTM) using MSE.

`init(self, season_length: int = 1, decomposition_type: str = 'multiplicative', model: str | None = None, alias: str = 'AutoTheta', prediction_intervals: ConformalIntervals | None = None, conformal_params: ConformalIntervals | None = None, n_samples: int = 200)`

(No prose summary provided in docstring.)

Parameter	Type	Default	Description
`season_length`	`int`	`1`	Number of observations per unit of time.
`decomposition_type`	`str`	`"multiplicative"`	Seasonal decomposition type: 'multiplicative' or 'additive'.
`model`	`str \\| None`	`None`	Controlling theta model variant. None searches the best model.
`alias`	`str`	`"AutoTheta"`	Custom name of the model.
`prediction_intervals`	`ConformalIntervals \\| None`	`None`	Configuration for conformal prediction intervals.
`conformal_params`	`ConformalIntervals \\| None`	`None`	Parameters for conformal prediction intervals.
`n_samples`	`int`	`200`	Number of Monte Carlo samples for prediction intervals.

`fit(self, y: jnp.ndarray, X: jnp.ndarray | None = None) -> Self`

Fit the AutoTheta model.

Parameter	Type	Default	Description
`y`	`jnp.ndarray`	-	Clean time series of shape (t,).
`X`	`jnp.ndarray \\| None`	`None`	Optional exogenous of shape (t, n_x).

Returns: Self (the fitted forecaster; sets self.model_). Raises: Exception

`predict(self, h: int, X: jnp.ndarray | None = None, level: list | None = None) -> dict`

Predict with fitted AutoTheta.

Parameter	Type	Default	Description
`h`	`int`	-	Forecast horizon.
`X`	`jnp.ndarray \\| None`	`None`	Optional exogenous of shape (h, n_x).
`level`	`list \\| None`	`None`	Confidence levels (0-100) for prediction intervals.

Returns: dict (Keys: 'mean' and optionally 'lo-{lv}', 'hi-{lv}').

`predict_in_sample(self, level: list | None = None) -> dict`

Access fitted AutoTheta in-sample predictions.

Parameter	Type	Default	Description
`level`	`list \\| None`	`None`	Confidence levels (0-100) for prediction intervals.

Returns: dict (Keys: 'fitted' and optionally 'lo-{lv}', 'hi-{lv}').

`forecast(self, y: jnp.ndarray, h: int, X: jnp.ndarray | None = None, X_future: jnp.ndarray | None = None, level: list | None = None, fitted: bool = False) -> dict`

Memory-efficient AutoTheta predictions. Fits and forecasts without storing model state.

Parameter	Type	Default	Description
`y`	`jnp.ndarray`	-	Clean time series of shape (t,).
`h`	`int`	-	Forecast horizon.
`X`	`jnp.ndarray \\| None`	`None`	Optional exogenous of shape (t, n_x).
`X_future`	`jnp.ndarray \\| None`	`None`	Optional future exogenous of shape (h, n_x).
`level`	`list \\| None`	`None`	Confidence levels (0-100) for prediction intervals.
`fitted`	`bool`	`False`	Whether to return in-sample predictions.

Returns: dict (Keys: 'mean', and optionally 'fitted', 'lo-{lv}', 'hi-{lv}').

`forward(self, y: jnp.ndarray, h: int, X: jnp.ndarray | None = None, X_future: jnp.ndarray | None = None, level: list | None = None, fitted: bool = False) -> dict`

Apply fitted AutoTheta model to a new time series. Uses the model type and parameters from the original fit.

Parameter	Type	Default	Description
`y`	`jnp.ndarray`	-	Clean time series of shape (n,).
`h`	`int`	-	Forecast horizon.
`X`	`jnp.ndarray \\| None`	`None`	Optional exogenous of shape (n, n_x).
`X_future`	`jnp.ndarray \\| None`	`None`	Optional future exogenous of shape (h, n_x).
`level`	`list \\| None`	`None`	Confidence levels (0-100) for prediction intervals.
`fitted`	`bool`	`False`	Whether to return in-sample predictions.

Returns: dict (Keys: 'mean', and optionally 'fitted', 'lo-{lv}', 'hi-{lv}'). Raises: Exception

`Theta`

chronax.models.Theta · inherits BaseForecaster

Standard Theta Method (STM). A simplified version of AutoTheta that always uses the Standard Theta Model.

`init(self, season_length: int = 1, decomposition_type: str = 'multiplicative', alias: str = 'Theta', prediction_intervals: ConformalIntervals | None = None, n_samples: int = 200)`

(No prose summary provided in docstring.)

Parameter	Type	Default	Description
`season_length`	`int`	`1`	Number of observations per unit of time.
`decomposition_type`	`str`	`"multiplicative"`	Seasonal decomposition type: 'multiplicative' or 'additive'.
`alias`	`str`	`"Theta"`	Custom name of the model.
`prediction_intervals`	`ConformalIntervals \\| None`	`None`	Configuration for conformal prediction intervals.
`n_samples`	`int`	`200`	Number of Monte Carlo samples for prediction intervals.

AutoTheta

__init__(self, season_length: int = 1, decomposition_type: str = 'multiplicative', model: str | None = None, alias: str = 'AutoTheta', prediction_intervals: ConformalIntervals | None = None, conformal_params: ConformalIntervals | None = None, n_samples: int = 200)

fit(self, y: jnp.ndarray, X: jnp.ndarray | None = None) -> Self

predict(self, h: int, X: jnp.ndarray | None = None, level: list | None = None) -> dict

predict_in_sample(self, level: list | None = None) -> dict

forecast(self, y: jnp.ndarray, h: int, X: jnp.ndarray | None = None, X_future: jnp.ndarray | None = None, level: list | None = None, fitted: bool = False) -> dict

forward(self, y: jnp.ndarray, h: int, X: jnp.ndarray | None = None, X_future: jnp.ndarray | None = None, level: list | None = None, fitted: bool = False) -> dict

Theta

__init__(self, season_length: int = 1, decomposition_type: str = 'multiplicative', alias: str = 'Theta', prediction_intervals: ConformalIntervals | None = None, n_samples: int = 200)