Skip navigation

Semi-parametric modelling of trend in extremes

Semi-parametric modelling of trend in extremes

Ramesh, N. I. ORCID: 0000-0001-6373-2557 and Davison, A. C. (2000) Semi-parametric modelling of trend in extremes. In: International Conference on Order Statistics & Extreme Values, 18-20 December 2000, Mysore, India. (Unpublished)

PDF (Author Accepted Abstract)
17985 RAMESH_Semi-Parametric_Modelling_2000.pdf - Presentation

Download (101kB) | Preview


Assessment of trend has been a topic of major interest in extreme value analysis in recent years. Much of the previous work has focused on using parametric techniques to model trend in extremes. The parametric approach, however, is often not flexible enough for exploratory modelling. In this paper, we discuss and illustrate a semi-parametric method which can be used as an exploratory tool to draw more information about the extremes and to model trend in extremes. Our approach is based on local likelihood fitting of generalized extreme value distribution and related models. It aims to capture the pattern of trend by fitting locally weighted polynomials to the model parameters. We illustrate the application of this methodology in a study investigating changes in extreme temperatures in central England. Bootstrap methods are used to provide a measure of the variability of the fitted quantities.

Item Type: Conference or Conference Paper (Paper)
Uncontrolled Keywords: Bootstrap confidence bands; Generalized extreme-value distribution; Local likelihood; Return level; Semi-parametric modelling; Trend analysis
Subjects: H Social Sciences > HA Statistics
Faculty / Department / Research Group: Faculty of Liberal Arts & Sciences
Faculty of Liberal Arts & Sciences > School of Computing & Mathematical Sciences (CAM)
Last Modified: 27 Oct 2020 14:50
Selected for GREAT 2016: None
Selected for GREAT 2017: None
Selected for GREAT 2018: None
Selected for GREAT 2019: None
Selected for REF2021: None

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics