/autoshiftedlog

Python Auto Shifted Log data transformation of Feng, Hannig, Marron

Primary LanguagePythonMIT LicenseMIT

Auto Shifted Log

The family of shifted log transformations helps correct nonnormality and skewness. The Automatic Shifted Log transformation of Feng, Hannig, and Marron automatically selects the shift parameter that minimizes either

  • the Anderson-Darling statistic (to correct nonnormality)
  • the sample skewness (to correct skewness).

Usage

from autoshiftedlog import autoshiftedlog

data = numpy.array([1, 1, 2, 2, 2.5, 3, 3, 4.5, 6, 8, 10, 20, 30, 80, 0, numpy.nan])
data = autoshiftedlog(data)