statsmodels.stats.diagnostic.compare_cox#

statsmodels.stats.diagnostic.compare_cox(results_x, results_z, store=False, *, result_object=None)[source]#

Compute the Cox test for non-nested models

Parameters:

results_xRegressionResults: The result instance of first model.
results_zRegressionResults: The result instance of second model.
storebool, default False: If true, then the intermediate results are returned.
result_objectbool, optional: Flag indicating whether to return the results as a NonNestedTestResult NamedTuple instead of a plain tuple. When store=True the NamedTuple holds the same three elements as the legacy tuple, so it unpacks identically and is always returned, with no warning. When store=False the legacy two-element tuple is returned by default and a FutureWarning is issued.

Deprecated since version 0.15.0: In release 0.16.0 or after July 2027, whichever is later, the default will change to always return a NonNestedTestResult. Set result_object=True to opt in now, or result_object=False to silence the warning and keep the current return type.

Returns:

NonNestedTestResult: If result_object=True, a NamedTuple with fields statistic, pvalue, and res_store (res_store is None when not computed). See NonNestedTestResult.
Otherwise (the deprecated default), a plain tuple made up of:
tstatfloat: t statistic for the test that including the fitted values of the first model in the second model has no effect.
pvaluefloat: two-sided pvalue for the t statistic
res_storeResultsStore, optional: Intermediate results. Returned if store is True.

Notes

Tests of non-nested hypothesis might not provide unambiguous answers. The test should be performed in both directions and it is possible that both or neither test rejects. See [1] for more information.

Formulas from [1], section 8.3.4 translated to code

Matches results for Example 8.3 in Greene

References

[1] (1,2)

Greene, W. H. Econometric Analysis. New Jersey. Prentice Hall; 5th edition. (2002).