A Noninformative Bayes-like Approach to Probability-Preserving Prediction of Extremes

2014/10/09 by Allan McRobie, McRobie, Allan
Computer Science · #Anomaly Detection Techniques and Applications #FOS: Mathematics #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.1410.2639

openalex publication_date 2014/10/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The extrapolation of extremes to values beyond the span of stationary univariate historical data is considered from Bayesian and Frequentist perspectives. The intention is to make predictions which in some sense "preserve probability". A Frequentist approach based on a simple curve-fit estimate of the tail parameter ξ of a Generalised Pareto Distribution was described in McRobie (2014) (arXiv:1408.1532). In this paper, the corresponding Bayes-like approach is described, using a plausible noninformative prior for the tail parameter. The two approaches, though philosophically different, show a reasonable degree of correspondence.

Citations

• An Intuitive Curve-Fit Approach to Probability-Preserving Prediction of\n Extremes

Related

• An Intuitive Curve-Fit Approach to Probability-Preserving Prediction of\n Extremes
• Bayesian prior elicitation and selection for extreme values
• Probability-Matching Predictors for Extreme Extremes
• Consistency of Bayesian Inference for Multivariate Max-Stable Distributions
• Statistical Prediction of Peaks Over a Threshold