Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

An extreme value approach to modelling number of causalities in earthquakes

Steneld, Henrik LU (2021) In Bachelor's Theses in Mathematical Sciences MASK11 20201
Mathematical Statistics
Abstract
Earthquakes occur around the globe all the time. Most are weak enough to just pass by, some are strong enough to be felt by us humans, and some very few are completely devastating. A comprehensive database distributed by NOAA National Centers for Environmental Information provides a means for reviewing devastating earthquakes over the past. Extreme value theory has previously been applied to modelling earthquakes, although for the most part the modelling has been concerned with the magnitudes. In this thesis, extreme value theory has been applied to the number of casualties that are directly or indirectly the result of an earthquake.

An in-homogeneous Poisson point process is fitted to events where the death toll is at least ten or... (More)
Earthquakes occur around the globe all the time. Most are weak enough to just pass by, some are strong enough to be felt by us humans, and some very few are completely devastating. A comprehensive database distributed by NOAA National Centers for Environmental Information provides a means for reviewing devastating earthquakes over the past. Extreme value theory has previously been applied to modelling earthquakes, although for the most part the modelling has been concerned with the magnitudes. In this thesis, extreme value theory has been applied to the number of casualties that are directly or indirectly the result of an earthquake.

An in-homogeneous Poisson point process is fitted to events where the death toll is at least ten or more. The events are assumed to be independent but non-stationary with respect to the magnitude of the earthquake. This leads to a Poisson point process with an intensity which is a function of magnitude. In addition, an assumption is made about the distribution of the magnitudes of earthquakes, which provides the necessary means for modelling extremes of earthquakes death toll unconditional of magnitude. With the aid of simulations and the asymptotic normality of maximum likelihood estimators, return levels and corresponding confidence intervals are calculated for three different geographical regions. (Less)
Please use this url to cite or link to this publication:
author
Steneld, Henrik LU
supervisor
organization
course
MASK11 20201
year
type
M2 - Bachelor Degree
subject
keywords
Extreme Value Theory, Point Process, Earthquakes
publication/series
Bachelor's Theses in Mathematical Sciences
report number
LUNFMS-4053-2021
ISSN
1654-6229
other publication id
2021:K11
language
English
id
9042231
date added to LUP
2021-05-12 10:01:53
date last changed
2021-06-03 15:08:52
@misc{9042231,
  abstract     = {{Earthquakes occur around the globe all the time. Most are weak enough to just pass by, some are strong enough to be felt by us humans, and some very few are completely devastating. A comprehensive database distributed by NOAA National Centers for Environmental Information provides a means for reviewing devastating earthquakes over the past. Extreme value theory has previously been applied to modelling earthquakes, although for the most part the modelling has been concerned with the magnitudes. In this thesis, extreme value theory has been applied to the number of casualties that are directly or indirectly the result of an earthquake.

An in-homogeneous Poisson point process is fitted to events where the death toll is at least ten or more. The events are assumed to be independent but non-stationary with respect to the magnitude of the earthquake. This leads to a Poisson point process with an intensity which is a function of magnitude. In addition, an assumption is made about the distribution of the magnitudes of earthquakes, which provides the necessary means for modelling extremes of earthquakes death toll unconditional of magnitude. With the aid of simulations and the asymptotic normality of maximum likelihood estimators, return levels and corresponding confidence intervals are calculated for three different geographical regions.}},
  author       = {{Steneld, Henrik}},
  issn         = {{1654-6229}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Bachelor's Theses in Mathematical Sciences}},
  title        = {{An extreme value approach to modelling number of causalities in earthquakes}},
  year         = {{2021}},
}