Natural (non-)informative priors for skew-symmetric distributions
- Author
- Holger Dette, Christophe Ley (UGent) and Francisco Rubio
- Organization
- Abstract
- In this paper, we present an innovative method for constructing proper priors for the skewness (shape) parameter in the skew-symmetric family of distributions. The proposed method is based on assigning a prior distribution on the perturbation effect of the shape parameter, which is quantified in terms of the total variation distance. We discuss strategies to translate prior beliefs about the asymmetry of the data into an informative prior distribution of this class. We show via a Monte Carlo simulation study that our non-informative priors induce posterior distributions with good frequentist properties, similar to those of the Jeffreys prior. Our informative priors yield better results than their competitors from the literature. We also propose a scale-invariant and location-invariant prior structure for models with unknown location and scale parameters and provide sufficient conditions for the propriety of the corresponding posterior distribution. Illustrative examples are presented using simulated and real data.
- Keywords
- BAYESIAN-ANALYSIS, T DISTRIBUTIONS, INFORMATION, STATISTICS, FAMILIES, measure of skewness, prior elicitation, skew-symmetric distributions, total variation distance
Downloads
-
A36 (DLR17).pdf
- full text
- |
- open access
- |
- |
- 607.85 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-8599278
- MLA
- Dette, Holger, et al. “Natural (Non-)Informative Priors for Skew-Symmetric Distributions.” SCANDINAVIAN JOURNAL OF STATISTICS, vol. 45, no. 2, 2018, pp. 405–20, doi:10.1111/sjos.12306.
- APA
- Dette, H., Ley, C., & Rubio, F. (2018). Natural (non-)informative priors for skew-symmetric distributions. SCANDINAVIAN JOURNAL OF STATISTICS, 45(2), 405–420. https://doi.org/10.1111/sjos.12306
- Chicago author-date
- Dette, Holger, Christophe Ley, and Francisco Rubio. 2018. “Natural (Non-)Informative Priors for Skew-Symmetric Distributions.” SCANDINAVIAN JOURNAL OF STATISTICS 45 (2): 405–20. https://doi.org/10.1111/sjos.12306.
- Chicago author-date (all authors)
- Dette, Holger, Christophe Ley, and Francisco Rubio. 2018. “Natural (Non-)Informative Priors for Skew-Symmetric Distributions.” SCANDINAVIAN JOURNAL OF STATISTICS 45 (2): 405–420. doi:10.1111/sjos.12306.
- Vancouver
- 1.Dette H, Ley C, Rubio F. Natural (non-)informative priors for skew-symmetric distributions. SCANDINAVIAN JOURNAL OF STATISTICS. 2018;45(2):405–20.
- IEEE
- [1]H. Dette, C. Ley, and F. Rubio, “Natural (non-)informative priors for skew-symmetric distributions,” SCANDINAVIAN JOURNAL OF STATISTICS, vol. 45, no. 2, pp. 405–420, 2018.
@article{8599278, abstract = {{In this paper, we present an innovative method for constructing proper priors for the skewness (shape) parameter in the skew-symmetric family of distributions. The proposed method is based on assigning a prior distribution on the perturbation effect of the shape parameter, which is quantified in terms of the total variation distance. We discuss strategies to translate prior beliefs about the asymmetry of the data into an informative prior distribution of this class. We show via a Monte Carlo simulation study that our non-informative priors induce posterior distributions with good frequentist properties, similar to those of the Jeffreys prior. Our informative priors yield better results than their competitors from the literature. We also propose a scale-invariant and location-invariant prior structure for models with unknown location and scale parameters and provide sufficient conditions for the propriety of the corresponding posterior distribution. Illustrative examples are presented using simulated and real data.}}, author = {{Dette, Holger and Ley, Christophe and Rubio, Francisco}}, issn = {{0303-6898}}, journal = {{SCANDINAVIAN JOURNAL OF STATISTICS}}, keywords = {{BAYESIAN-ANALYSIS,T DISTRIBUTIONS,INFORMATION,STATISTICS,FAMILIES,measure of skewness,prior elicitation,skew-symmetric distributions,total variation distance}}, language = {{eng}}, number = {{2}}, pages = {{405--420}}, title = {{Natural (non-)informative priors for skew-symmetric distributions}}, url = {{http://doi.org/10.1111/sjos.12306}}, volume = {{45}}, year = {{2018}}, }
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: