`enc_t` – Enc-T forecast encompassing test (Ericsson 1992 t-form).#

Back to nested_test axis | Back to L6 | Browse all options

Operational op under axis nested_test, sub-layer L6_B_nested, layer l6. Standalone callable: mf.functions.enc_t_test.

Function signature#

mf.functions.enc_t_test(
    loss_small: np.ndarray,
    loss_large: np.ndarray,
) -> EncTTestResult

Parameters#

name	type	default	constraint	description
`loss_small`	`np.ndarray`	—	—	Squared losses for the small model.
`loss_large`	`np.ndarray`	—	—	Squared losses for the large model.

Returns#

EncTTestResult — frozen dataclass with fit results.

Attribute	Type	Description
`stat`	`float or None`	Enc-T statistic
`pvalue`	`float or None`	One-sided p-value
`decision`	`bool`	Reject H0 at 5%
`n_obs`	`int`	Observations used

Behavior#

Ericsson (1992) t-form of the encompassing test. Identical computation to enc_new in the current implementation (raw loss improvement, one-sided DM inference, no CW adjustment). The distinction is the conceptual labelling: enc_t is cast as a t-statistic on the mean loss improvement. Both enc_new and enc_t share the same runtime dispatch branch.

When to use

Encompassing tests in contexts where the Ericsson t-form labelling is preferred.

When NOT to use

When CW adjustment is needed – use clark_west instead.

In recipe context#

Set params.nested_test = "enc_t" in the relevant layer to activate this op within a recipe:

# Layer L6 recipe fragment
params:
  nested_test: enc_t

References#

macroforecast design Part 3, L6: ‘tests must report (statistic, p-value, kernel, lag) and respect HAC dependence-correction.’
Ericsson (1992) ‘Parameter Constancy, Mean Square Forecast Errors, and Measuring Forecast Performance’, JoE 52(1-2): 113-153.

enc_t – Enc-T forecast encompassing test (Ericsson 1992 t-form).#