Naive

chronax.models.naive.Naive · inherits BaseForecaster

The naive class implements statsforecast's naive forecasting model, in which any forecast is equal to the previously observed value. This JAX implementation provides complete compatibility with statsforecast's Naive class, including: - fit(): Trains the model and stores parameters - predict(): Makes forecasts using fitted model (stateful) with confidence intervals - predict_in_sample(): Returns fitted values from training with confidence intervals - forecast(): Stateless prediction (fit + predict in one step) with confidence intervals - forward(): Applies fitted model to new/updated time series Core computation is handled by _naive_core() which computes forecasts, fitted values, and residual statistics. Confidence Intervals: - Native intervals: Uses residual standard error with normal distribution assumption - Conformal intervals: Uses base_forecaster's conformal prediction methods when conformal_params is provided

__init__(self, alias: str = 'Naive', conformal_params: ConformalIntervals | None = None)

(Initializes the Naive forecaster.)

Parameter	Type	Default	Description
alias	str	"Naive"	(undocumented)
conformal_params	ConformalIntervals \| None	None	(undocumented)

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

Trains the model and stores parameters.

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

Predict with fitted Naive.

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

Returns: dict (Dictionary with entries mean for point predictions and level_* for probabilistic predictions.)

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

Access fitted Naive insample predictions.

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

Returns: dict (Dictionary with entries fitted for point predictions.)

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

Memory Efficient Naive predictions.

This method avoids memory burden due from object storage. It is analogous to fit_predict without storing information. It assumes you know the forecast horizon in advance.

Parameter	Type	Default	Description
h	int	-	Forecast horizon.
y	jnp.ndarray	-	Clean time series of shape (n,).
X	jnp.ndarray \| None	None	Optional insample exogenous of shape (t, n_x).
X_future	jnp.ndarray \| None	None	Optional exogenous of shape (h, n_x).
level	list[int] \| None	None	Confidence levels (0-100) for prediction intervals.
fitted	bool	False	Whether or not to return insample predictions.

Returns: dict (Dictionary with entries mean for point predictions and level_* for probabilistic predictions.)

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

Apply fitted model to an new/updated series.

Parameter	Type	Default	Description
h	int	-	Forecast horizon.
y	jnp.ndarray	-	Clean time series of shape (n,).
X	jnp.ndarray \| None	None	Optional insample exogenous of shape (t, n_x).
X_future	jnp.ndarray \| None	None	Optional exogenous of shape (h, n_x).
level	list[int] \| None	None	Confidence levels (0-100) for prediction intervals.
fitted	bool	False	Whether or not to return insample predictions.

Returns: dict (Dictionary with entries mean for point predictions and level_* for probabilistic predictions.)