gw_giacomini_white – Giacomini-White (2006) conditional equal-predictive-ability test.#

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

Operational op under axis equal_predictive_test, sub-layer L6_A_equal_predictive, layer l6. Standalone callable: mf.functions.gw_test.

Function signature#

mf.functions.gw_test(
    loss_a: np.ndarray,
    loss_b: np.ndarray,
) -> GWTestResult

Parameters#

name

type

default

constraint

description

loss_a

np.ndarray

—

—

Per-period losses for model A (e.g. squared errors).

loss_b

np.ndarray

—

—

Per-period losses for model B.

Returns#

GWTestResult — frozen dataclass with fit results.

Attribute

Type

Description

stat

float or None

GW test statistic

pvalue

float or None

Two-sided p-value

decision

bool

Reject H0 at 5%

n_obs

int

Observations used

hln_correction

bool

HLN correction applied

Behavior#

Generalises DM to test conditional predictive ability given a vector of predictors. Robust to non-stationary performance differentials and works with rolling / expanding-window forecasts.

When to use

Conditional / regime-dependent forecast comparisons.

In recipe context#

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

# Layer L6 recipe fragment
params:
  equal_predictive_test: gw_giacomini_white

References#

  • macroforecast design Part 3, L6: ‘tests must report (statistic, p-value, kernel, lag) and respect HAC dependence-correction.’

  • Giacomini & White (2006) ‘Tests of Conditional Predictive Ability’, Econometrica 74(6): 1545-1578.