A new weighted quantile regression

The objective of the study is to use quantile regression to estimate extreme value events. The exploration of extreme value events requires the use of heavy-tailed distributions to build a model which fits the data well. One needs to estimate high conditional quantiles of a random variable for extre...

Full description

Bibliographic Details
Main Authors: Mei Ling Huang, Ramona Rat
Format: Article
Language:English
Published: Taylor & Francis Group 2017-01-01
Series:Cogent Mathematics
Subjects:
Online Access:http://dx.doi.org/10.1080/23311835.2017.1357237
id doaj-e235575bca2a4c7e86fff5a72d4ea22c
record_format Article
spelling doaj-e235575bca2a4c7e86fff5a72d4ea22c2020-11-24T21:34:58ZengTaylor & Francis GroupCogent Mathematics2331-18352017-01-014110.1080/23311835.2017.13572371357237A new weighted quantile regressionMei Ling Huang0Ramona Rat1Brock UniversityBrock UniversityThe objective of the study is to use quantile regression to estimate extreme value events. The exploration of extreme value events requires the use of heavy-tailed distributions to build a model which fits the data well. One needs to estimate high conditional quantiles of a random variable for extreme events. Quantile regression ultimately yields results which the alternative mean regression method has no hope to offer, leading to it being labeled as the more powerful method. In order to improve this approach even further, a weighted quantile regression method is introduced with a complete comparison to the unweighted method. The Monte Carlo simulations show good results for the proposed weighted method. Comparisons of the proposed method and existing methods are given. The paper also investigates two real-world examples of applications on extreme events using the proposed weighted method.http://dx.doi.org/10.1080/23311835.2017.1357237bivariate Pareto distribution Type IIconditional quantilekernel conditional density estimatorgeneralized Pareto distributionlinear programmingweighted loss function
collection DOAJ
language English
format Article
sources DOAJ
author Mei Ling Huang
Ramona Rat
spellingShingle Mei Ling Huang
Ramona Rat
A new weighted quantile regression
Cogent Mathematics
bivariate Pareto distribution Type II
conditional quantile
kernel conditional density estimator
generalized Pareto distribution
linear programming
weighted loss function
author_facet Mei Ling Huang
Ramona Rat
author_sort Mei Ling Huang
title A new weighted quantile regression
title_short A new weighted quantile regression
title_full A new weighted quantile regression
title_fullStr A new weighted quantile regression
title_full_unstemmed A new weighted quantile regression
title_sort new weighted quantile regression
publisher Taylor & Francis Group
series Cogent Mathematics
issn 2331-1835
publishDate 2017-01-01
description The objective of the study is to use quantile regression to estimate extreme value events. The exploration of extreme value events requires the use of heavy-tailed distributions to build a model which fits the data well. One needs to estimate high conditional quantiles of a random variable for extreme events. Quantile regression ultimately yields results which the alternative mean regression method has no hope to offer, leading to it being labeled as the more powerful method. In order to improve this approach even further, a weighted quantile regression method is introduced with a complete comparison to the unweighted method. The Monte Carlo simulations show good results for the proposed weighted method. Comparisons of the proposed method and existing methods are given. The paper also investigates two real-world examples of applications on extreme events using the proposed weighted method.
topic bivariate Pareto distribution Type II
conditional quantile
kernel conditional density estimator
generalized Pareto distribution
linear programming
weighted loss function
url http://dx.doi.org/10.1080/23311835.2017.1357237
work_keys_str_mv AT meilinghuang anewweightedquantileregression
AT ramonarat anewweightedquantileregression
AT meilinghuang newweightedquantileregression
AT ramonarat newweightedquantileregression
_version_ 1725947089997791232