Rights statement: The final, definitive version of this article has been published in the Journal, Technometrics, ?? (?), 2015, © Informa Plc
Accepted author manuscript, 613 KB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Exploiting structure of maximum likelihood estimators for extreme value threshold selection
AU - Wadsworth, Jennifer
N1 - The final, definitive version of this article has been published in the Journal, Technometrics, 58 (1), 2016, © Informa Plc
PY - 2016/3
Y1 - 2016/3
N2 - In order to model the tail of a distribution, one has to define the threshold above or below which an extreme value model produces a suitable fit. Parameter stability plots, whereby one plots maximum likelihood estimates of supposedly threshold-independent parameters against threshold, form one of the main tools for threshold selection by practitioners, principally due to their simplicity. However, one repeated criticism of these plots is their lack of interpretability, with pointwise confidence intervals being strongly dependent across the range of thresholds. In this article, we exploit the independent-increments structure of maximum likelihood estimators in order to produce complementary plots with greater interpretability, and a suggest simple likelihood-based procedure which allows for automated threshold selection.
AB - In order to model the tail of a distribution, one has to define the threshold above or below which an extreme value model produces a suitable fit. Parameter stability plots, whereby one plots maximum likelihood estimates of supposedly threshold-independent parameters against threshold, form one of the main tools for threshold selection by practitioners, principally due to their simplicity. However, one repeated criticism of these plots is their lack of interpretability, with pointwise confidence intervals being strongly dependent across the range of thresholds. In this article, we exploit the independent-increments structure of maximum likelihood estimators in order to produce complementary plots with greater interpretability, and a suggest simple likelihood-based procedure which allows for automated threshold selection.
KW - diagnostic plots
KW - extreme value modelling
KW - maximum likelihood
KW - threshold selection
U2 - 10.1080/00401706.2014.998345
DO - 10.1080/00401706.2014.998345
M3 - Journal article
VL - 58
SP - 116
EP - 126
JO - Technometrics
JF - Technometrics
SN - 0040-1706
IS - 1
ER -