def quantile_CI()

in bayesmark/quantiles.py [0:0]

• 10 lines of code
• 5 McCabe index (conditional complexity)

def quantile_CI(X, q, alpha=0.05):
    """Calculate CI on `q` quantile from same `X` using nonparametric estimation from order statistics.

    This will have alpha level of at most `alpha` due to the discrete nature of order statistics.

    Parameters
    ----------
    X : :class:`numpy:numpy.ndarray` of shape (n,)
        Data for quantile estimation. Can be vectorized. Must be sortable data type (which is almost everything).
    q : float
        Quantile to compute, must be in (0, 1). Can be vectorized.
    alpha : float
        False positive rate we allow for CI, must be in (0, 1). Can be vectorized.

    Returns
    -------
    LB : dtype of `X`, scalar
        Lower end on CI
    UB : dtype of `X`, scalar
        Upper end on CI
    """
    assert np.ndim(X) >= 1
    # We could robustify things to allow the edge cases, but maybe later
    assert np.all(0 < q) and np.all(q < 1)
    assert np.all(0 < alpha) and np.all(alpha < 1)
    # Currently don't support broadcasting both at same time
    assert np.ndim(X) == 1 or (np.ndim(q) == 0 and np.ndim(alpha) == 0)

    n = X.shape[-1]
    idx_lower, idx_upper = _quantile_CI(n, q, alpha)

    o_stats = order_stats(X)
    LB, UB = o_stats[..., idx_lower], o_stats[..., idx_upper]
    return LB, UB