EconStor >
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) >
Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung (IWQW), Universität Erlangen-Nürnberg >
IWQW Discussion Paper Series, FAU Erlangen-Nürnberg >

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10419/52731
  
Title:Van Zwet ordering and the Ferreira-Steel family of skewed distributions PDF Logo
Authors:Klein, Ingo
Issue Date:2011
Series/Report no.:IWQW discussion paper series 13/2011
Abstract:There are several procedures to construct a skewed distribution. One of these procedures is based on a symmetric distribution that will be distorted by a skewed distribution defined on (0; 1). This proposal stems from Arellano-Valle et al. and was refined by Ferreira and Steel. Up to now, it is an open question whether the famous skewness ordering of van Zwet will be preserved for this proposal. There is a general condition under which the van Zwets skewness ordering will be preserved by the Ferreira-Steel family. But this condition is not easy to verify for the most families of distribution. Therefore, for the skewness mechanism we choose a special beta distribution with only one parameter. Then, we get three results. First, the skewness ordering will be preserved starting for symmetric distributions that are leptokurtic like the logistic distribution. Larger parameter values give distributions that are more skewed to the right. Second, the same skewness mechanism can generate distributions that are more skewed to left if the support of the underlying symmetric distribution is compact. Third, for underlying symmetric distributions on R with platykurtic behavior the van Zwet ordering of skewness will be preserved. This restricts a little bit the benefit of the Ferreira-Steel family.
Subjects:skewness
skewness to the right
skewness ordering
measure of skewness
Document Type:Working Paper
Appears in Collections:IWQW Discussion Paper Series, FAU Erlangen-Nürnberg

Files in This Item:
File Description SizeFormat
678360928.pdf518.32 kBAdobe PDF
No. of Downloads: Counter Stats
Download bibliographical data as: BibTeX
Share on:http://hdl.handle.net/10419/52731

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.