plot_forecast

chronax.plotting.plot_forecast

Plots the actual ground truth values against the forecasted values.

plot_forecast(y, y_hat, y_train=None, ax=None) -> Any

Parameter	Type	Default	Description
y	jnp.ndarray	-	The actual time series values for the forecast horizon.
y_hat	jnp.ndarray	-	The predicted forecast values.
y_train	Optional[jnp.ndarray]	None	Historical training data to plot before the forecast.
ax	Optional[Any]	None	A matplotlib Axes object to plot on. If None, a new figure is created.

Returns: Any (The matplotlib Axes object containing the generated plot.)

plot_forecast_intervals

chronax.plotting.plot_forecast_intervals

Plots the forecast along with shaded prediction intervals.

plot_forecast_intervals(y, y_hat, lower, upper, ax=None) -> Any

Parameter	Type	Default	Description
y	jnp.ndarray	-	The actual time series values.
y_hat	jnp.ndarray	-	The predicted mean/median forecast values.
lower	jnp.ndarray	-	The lower bounds of the prediction interval.
upper	jnp.ndarray	-	The upper bounds of the prediction interval.
ax	Optional[Any]	None	A matplotlib Axes object.

Returns: Any (The matplotlib Axes object containing the generated plot.)

plot_forecast_distribution

chronax.plotting.plot_forecast_distribution

Plots a fan chart showing the forecast distribution using shaded percentile bands.

plot_forecast_distribution(y, y_samples, ax=None, percentiles=(10, 25, 50, 75, 90, 95)) -> Any

Parameter	Type	Default	Description
y	jnp.ndarray	-	The actual time series values.
y_samples	jnp.ndarray	-	A 2D array of simulated forecast samples (shape: [num_samples, horizon]).
ax	Optional[Any]	None	A matplotlib Axes object.
percentiles	Tuple[int, ...]	(10, 25, 50, 75, 90, 95)	The specific percentiles to plot as shaded regions.

Returns: Any (The matplotlib Axes object containing the generated fan chart.)

plot_forecast_pdf

chronax.plotting.plot_forecast_pdf

Plots the Probability Density Function (PDF) of the forecast samples at a specific horizon step.

plot_forecast_pdf(y_samples, horizon_idx=-1, bins=30, ax=None) -> Any

Parameter	Type	Default	Description
y_samples	jnp.ndarray	-	A 2D array of simulated forecast samples.
horizon_idx	int	-1	The specific index of the forecast horizon to plot. Defaults to -1 (the final step).
bins	int	30	The number of bins to use for the histogram.
ax	Optional[Any]	None	A matplotlib Axes object.

Returns: Any (The matplotlib Axes object containing the plotted PDF.)

plot_chained_window

chronax.plotting.plot_chained_window

Plots sequential cross-validation windows to evaluate rolling model performance.

plot_chained_window(y, y_preds, horizon=1, ax=None) -> Any

Parameter	Type	Default	Description
y	jnp.ndarray	-	The actual continuous time series values.
y_preds	List[jnp.ndarray]	-	A list of forecasted arrays, each representing a CV window.
horizon	int	1	The forecast horizon length (currently unused in logic, reserved for future functionality).
ax	Optional[Any]	None	A matplotlib Axes object.

Returns: Any (The matplotlib Axes object containing the chained cross-validation plot.)

acf

chronax.plotting.acf

Computes the Autocorrelation Function (ACF) array for a given time series.

acf(x, nlags=40) -> jnp.ndarray

Parameter	Type	Default	Description
x	jnp.ndarray	-	The input time series data.
nlags	int	40	The maximum number of lags to compute.

Returns: jnp.ndarray (An array containing the autocorrelation coefficients from lag 0 up to nlags.)

plot_acf

chronax.plotting.plot_acf

Computes and plots the Autocorrelation Function (ACF) along with 95% statistical significance bounds.

plot_acf(x, nlags=40) -> None

Parameter	Type	Default	Description
x	jnp.ndarray	-	The input time series data.
nlags	int	40	The number of lags to compute and display.

Returns: None (Displays the plot via plt.show().)