TSB

chronax.tsb.TSB · inherits BaseForecaster

Implements the Time Series Buffer (TSB) algorithm, applying exponential smoothing to both demand probability and demand magnitude, suitable for intermittent time series.

Attributes: * uses_exog: bool = False * alias: str * conformal_params: ConformalIntervals | None * model_: dict | None (Stores fitted parameters, including demand_level, prob_level, and sigma.)

__init__(self, alpha_d: float, alpha_p: float, alias: str = 'TSB', conformal_params: ConformalIntervals | None = None)

Parameter	Type	Default	Description
alpha_d	float	-	(undocumented)
alpha_p	float	-	(undocumented)
alias	str	"TSB"	(undocumented)
conformal_params	ConformalIntervals \| None	None	(undocumented)

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

Parameter	Type	Default	Description
y	jnp.ndarray	-	(undocumented)
X	jnp.ndarray \| None	None	(undocumented)

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

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

Parameter	Type	Default	Description
h	int	-	(undocumented)
X	jnp.ndarray \| None	None	(undocumented)
level	list[int] \| None	None	(undocumented)

Returns: dict

Key	Type	Description
mean	jnp.ndarray	The point forecasts of shape (h,).
lo-L	jnp.ndarray	Lower bound for confidence level L (if level is provided).
hi-L	jnp.ndarray	Upper bound for confidence level L (if level is provided).

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

Parameter	Type	Default	Description
level	list[int] \| None	None	(undocumented)

Returns: dict

Key	Type	Description
fitted	jnp.ndarray	The in-sample fitted values.
fitted-lo-L	jnp.ndarray	Lower bound for fitted confidence level L (if level is provided).
fitted-hi-L	jnp.ndarray	Upper bound for fitted confidence level L (if level is provided).

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

Parameter	Type	Default	Description
y	jnp.ndarray	-	(undocumented)
h	int	-	(undocumented)
X	jnp.ndarray \| None	None	(undocumented)
X_future	jnp.ndarray \| None	None	(undocumented)
level	list[int] \| None	None	(undocumented)
fitted	bool	False	(undocumented)

Returns: dict

Key	Type	Description
mean	jnp.ndarray	The point forecasts of shape (h,).
fitted	jnp.ndarray	The in-sample fitted values (if fitted=True).
lo-L	jnp.ndarray	Lower bound for confidence level L (if level is provided).
hi-L	jnp.ndarray	Upper bound for confidence level L (if level is provided).
fitted-lo-L	jnp.ndarray	Lower bound for fitted confidence level L (if level and fitted=True are provided).
fitted-hi-L	jnp.ndarray	Upper bound for fitted confidence level L (if level and fitted=True are provided).

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

(This method is an alias for forecast.)

Parameter	Type	Default	Description
y	jnp.ndarray	-	(undocumented)
h	int	-	(undocumented)
X	jnp.ndarray \| None	None	(undocumented)
X_future	jnp.ndarray \| None	None	(undocumented)
level	list[int] \| None	None	(undocumented)
fitted	bool	False	(undocumented)

Returns: The result of self.forecast.