Statistics

z_score_from_confidence_interval(confidence)

Given a confidence interval, this returns the z-score, \(Z\). Similar to obtaining z-score from P-value with a double tail distribution.

If variable, \(x\), has standard deviation, \(\sigma\), \(x \pm Z \sigma\) is the value of \(x\) with the desired uncertainty.

Parameters:

Name	Type	Description	Default
confidence	Union[array, float]	The confidence interval (between 0 and 1) required, e.g. 0.95 for 95% confidence, will return a z-score of 1.96.	required

Returns:

Type	Description
Union[array, float]	Z-score corresponding to confidence interval

Source code in isca_tools/utils/stats.py

def z_score_from_confidence_interval(confidence: Union[np.array, float]) -> Union[np.array, float]:
    """
    Given a confidence interval, this returns the z-score, $Z$. Similar to obtaining z-score from P-value with a double
    tail distribution.

    If variable, $x$, has standard deviation, $\sigma$, $x \pm Z \sigma$ is the value of $x$ with the
    desired uncertainty.

    Args:
        confidence: The confidence interval (between 0 and 1) required,
            e.g. 0.95 for 95% confidence, will return a z-score of 1.96.

    Returns:
        Z-score corresponding to confidence interval
    """
    return scipy.stats.norm.ppf(1-(1-confidence)/2)