climpred.classes.PerfectModelEnsemble.bootstrap

PerfectModelEnsemble.bootstrap(metric=None, comparison=None, dim=None, reference=None, iterations=None, sig=95, pers_sig=None, **metric_kwargs)[source]

Bootstrap with replacement according to Goddard et al. 2013.

Parameters

  • metric (str, Metric) – Metric to verify bootstrapped skill, see metrics.

  • comparison (str, Comparison) – Comparison passed to verify, see comparisons.

  • dim (str, list of str) – Dimension(s) over which to apply metric. dim is passed on to xskillscore.{metric} and includes xskillscore’s member_dim. dim should contain member when comparison is probabilistic but should not contain member when comparison=e2c. Defaults to None meaning that all dimensions other than lead are reduced.

  • reference (str, list of str) – Type of reference forecasts with which to verify. One or more of [‘persistence’, ‘uninitialized’]. If None or empty, returns no p value.

  • iterations (int) – Number of resampling iterations for bootstrapping with replacement. Recommended >= 500.

  • sig (int, default 95) – Significance level in percent for deciding whether uninitialized and persistence beat initialized skill.

  • pers_sig (int) – If not None, the separate significance level for persistence. Defaults to None, or the same significance as sig.

  • **metric_kwargs (optional) – arguments passed to metric.

Returns

with dimensions result (holding verify skill, p, low_ci and high_ci) and skill (holding initialized, persistence and/or uninitialized):

  • result=’verify skill’, skill=’initialized’:

    mean initialized skill

  • result=’high_ci’, skill=’initialized’:

    high confidence interval boundary for initialized skill

  • result=’p’, skill=’uninitialized’:

    p value of the hypothesis that the difference of skill between the initialized and uninitialized simulations is smaller or equal to zero based on bootstrapping with replacement.

  • result=’p’, skill=’persistence’:

    p value of the hypothesis that the difference of skill between the initialized and persistenceistence simulations is smaller or equal to zero based on bootstrapping with replacement.

Return type

xr.Datasets

Reference:

  • Goddard, L., A. Kumar, A. Solomon, D. Smith, G. Boer, P. Gonzalez, V. Kharin, et al. “A Verification Framework for Interannual-to-Decadal Predictions Experiments.” Climate Dynamics 40, no. 1–2 (January 1, 2013): 245–72. https://doi.org/10/f4jjvf.