Abstract
In bayesian inference, the Bayes estimator is the alternative with the minimum expected loss. In most cases, the loss function shows the distance between the alternative and the parameter. Therefore, any distance can lead to a loss function. Among the best known distance functions is L p one, where the choice of value p may be difficult and arbitrary. This paper examines robust models where the loss function is modelled by family L p . Our solution concept is the non-dominated alternative. We characterize the non-dominated set by having the posterior distribution function satisfy a particular asymmetry property. We also include an example to illustrate the methodology described.
This is a preview of subscription content, log in via an institution to check access.
We’re sorry, something doesn't seem to be working properly.
Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.
References
Arias-Nicolás JP and Martín J (2002). Uncertainty in beliefs and preferences: Conditions for optimal alternatives. Ann Math Artif Intell 35: 3–10
Arias-Nicolás JP, Martín J and Suárez-Llorens A (2006). The non-dominated set in bayesian decision problems with convex loss functions. Communications in Statistics. Theory Methods 35: 593–607
Berger J (1994). An overview of robust Bayesian analysis (with discussion). Test 3: 5–124
Berger J et al (1996) Bayesian Robustness. IMS lecture notes-monograph series, vol 29. IMS, Hayward
Dey D, Lou K and Bose S (1998). A Bayesian approach to loss robustness. Stat Decis 16: 65–87
Doksum KA (1975). Measures of location and asymmetry. Scand J Stat 2: 11–22
Makov UE (1994). Some aspects of Bayesian loss robustness. J Stat Plan Inference 38: 359–370
Martín J, Ríos Insua D and Ruggeri F (1998). Issues in Bayesian Loss Robustness. Sankhya Indian J Stat Ser A 60: 405–417
Moreno E (2000). Global Bayesian robustness for some classes of prior distributions. In: Ruggeri, F and Ríos Insua, D (eds) Robust Bayesian analysis, pp. Springer, Heidelberg
Ríos Insua D and Ruggeri F (2000). Bayesian sensitivity analysis. Springer, New York
Ruggeri F, Ríos Insua D and Martín J (2005). Robust Bayesian analysis. In: Dey, D and Rao, CR (eds) Handbook of statistics, vol 25 Bayesian thinking, modeling and computation., pp. North-Holland, Amsterdam
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Arias-Nicolás, J.P., Martín, J. & Suárez-Llorens, A. \({\mathcal{L}}_p\) loss functions: a robust bayesian approach. Stat Papers 50, 501–509 (2009). https://doi.org/10.1007/s00362-007-0093-0
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-007-0093-0