pylimma.zscore_t

pylimma.zscore_t(x, df, approx=False, method='bailey')[source]

Z-score equivalents of t-statistics.

Port of R limma’s zscoreT. When approx=True the calculation uses one of three closed-form approximations selected by method ("bailey", "hill", "wallace"). When approx=False the z-scores are computed via the exact quantile transformation using the log-scale t-distribution survival function; method is ignored on that path.

Parameters:

  • x (array_like) – t-statistics (any shape).

  • df (array_like or float) – Degrees of freedom. Broadcasts against x.

  • approx (bool, default False) – If True, use a closed-form approximation.

  • method ({"bailey", "hill", "wallace"}, default "bailey") – Approximation to use when approx=True. Ignored when approx=False.

Returns:

Z-score equivalents with the same shape as x.

Return type:

ndarray