A new family of power transformations to improve normality or symmetry

By I. Yeo, Richard A. Johnson. 2000

1,605 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Developmental and Educational Psychology, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/b38a5f78f6f5e1b9cac33c69c8ee1288b5d115ec

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Developmental and Educational Psychology

In Management Science and Operations Research

Citations in the corpus

Finding Optimal Normalizing Transformations via bestNormalize (2021, methodology): “This could be something as simple as a log transformation, or something as complex as a Yeo-Johnson transformation (Yeo and Johnson, 2000).”
A Modified Family of Power Transformations (2006, methodology): “To overcome this problem, many alternative transformations have been proposed, including the most recent one by Yeo and Johnson (2000).”
Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions (2019, background): “s in y t 1. Thus, the shape of the data is not changed but the series is simply shifted down by 1 (Rob J Hyndman,2018). Another similar form of power transformation is the Yeo-Johnson transformation (Yeo and Johnson, 2000) shown in Equation12. w t= 8 >> >< >> >: ((y t+ 1) 1)= if”
A Bayesian joint probability modeling approach for seasonal forecasting of streamflows at multiple sites (2009, methodology): “Parameter s(2) is an approximation of the variance of y [Yeo and Johnson, 2000].”
The neglog transformation and quantile regression for the analysis of a large credit scoring database (2005, background): “We show that the neglog transformation for variables that may be substantially negative, a relation of the Yeo–Johnson family of transformations (Yeo and Johnson, 2000), fulfils this aim on this database and so provides a convincing real world example of its use.”
Multisite probabilistic forecasting of seasonal flows for streams with zero value occurrences (2011, methodology): “, d, m is an approximation of the mean of a nontransformed variable y, and s is an approximation of the standard deviation of y [Yeo and Johnson, 2000].”
Merging Seasonal Rainfall Forecasts from Multiple Statistical Models through Bayesian Model Averaging (2012, methodology): “Given a predictor vector y(1) and a predictand vector y(2), Yeo–Johnson transforms (Yeo and Johnson 2000) are applied to each of the predictor and predictand variables.”