Supplementary material to the paper
| Anthony C. Atkinson | Marco Riani | Aldo Corbellini |
| The London School of Economics, | Department of Economics and Managemenet and Interdepartmental Centre for Robust Statistics | Department of Economics and Managemenet and Interdepartmental Centre for Robust Statistics |
| London WC2A 2AE, UK | University of Parma | University of Parma |
| UK | Italy | Italy |
| a.c.atkinson@lse.ac.uk | mriani@unipr.it | aldo.corbellini@unipr.it |
Abstract
The Yeo-Johnson transformation extends
the Box-Cox transformation for responses in linear models to observations that can
be positive or negative. The extended Yeo-Johnson transformation
allows positive and negative responses to have different transformations. Analyses
of data show this to be necessary. Robustness enters in the fan plot for which the
forward search provides an ordering of the data. Plausible transformations are checked
with an extended fan plot. These procedures are used to compare parametric power
transformations with the non-parametric transformations produced by ACE and AVAS.
We suggest further directions for this kind of research and, on the way, mention
a little history.
Data used in the paper are included in release 2019A of the FSDA toolbox downloadable from http://rosa.unipr.it/FSDA/fsdadownload.html
Matlab code used in the paper
The new routines are part of the MATLAB FSDA toolbox which can be downloaded from http://rosa.unipr.it
The link to the documentation of all these functions is at http://rosa.unipr.it/FSDA/function-cate.html
The matlab file which creates all the figures related to the gasoline data can be downloaded here
The matlab file which creates all the figures related to poison data can be downloaded here
The matlab file which creates all the figures related to simulated data can be downloaded here
The matlab file which creates all the figures related to the balance sheets data can be downloaded here
The matlab file which creates all the figures related to John and Draper difference data can be downloaded here
Last modified 20/05/2019 12.09.16