arch.bootstrap.SPA.better_models¶

SPA.better_models(pvalue: float = 0.05, pvalue_type: 'lower' | 'consistent' | 'upper' = 'consistent') → ndarray | list[Hashable][source]¶

Returns set of models rejected as being equal-or-worse than the benchmark

Parameters:¶

pvalue: float = 0.05¶

P-value in (0,1) to use when computing superior models

pvalue_type: 'lower' | 'consistent' | 'upper' = 'consistent'¶

String in ‘lower’, ‘consistent’, or ‘upper’ indicating which critical value to use.

Returns:¶

indices – List of column names or indices of the superior models. Column names are returned if models is a DataFrame.

Return type:¶

list

Notes

List of superior models returned is always with respect to the initial set of models, even when using subset().