Robust and Efficient Estimation for the Generalized Pareto Distribution View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2004-09

AUTHORS

Sergio F. Juárez, William R. Schucany

ABSTRACT

In this article we implement the minimum density power divergence estimator (MDPDE) for the shape and scale parameters of the generalized Pareto distribution (GPD). The MDPDE is indexed by a constant α ≥ 0 that controls the trade-off between robustness and efficiency. As α increases, robustness increases and efficiency decreases. For α = 0 the MDPDE is equivalent to the maximum likelihood estimator (MLE). We show that for α > 0 the MDPDE for the GPD has a bounded influence function. For α < 0.2 the MDPDE maintains good asymptotic relative efficiencies, usually above 90%. The results from a Monte Carlo study agree with these asymptotic calculations. The MDPDE is asymptotically normally distributed if the shape parameter is less than (1 + α)/(2 + α), and estimators for standard errors are readily computed under this restriction. We compare the MDPDE, MLE, Dupuis’ optimally-biased robust estimator (OBRE), and Peng and Welsh’s Medians estimator for the parameters. The simulations indicate that the MLE has the highest efficiency under uncontaminated GPDs. However, for the GPD contaminated with gross errors OBRE and MDPDE are more efficient than the MLE. For all the simulated models that we studied the Medians estimator had poor performance. More... »

PAGES

237-251

References to SciGraph publications

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s10687-005-6475-6

DOI

http://dx.doi.org/10.1007/s10687-005-6475-6

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1034725432


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Faculty of Statistics and Informatics, Veracruzana University, Xalapa, Ver, Mexico", 
          "id": "http://www.grid.ac/institutes/grid.42707.36", 
          "name": [
            "Faculty of Statistics and Informatics, Veracruzana University, Xalapa, Ver, Mexico"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ju\u00e1rez", 
        "givenName": "Sergio F.", 
        "id": "sg:person.012227626342.30", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012227626342.30"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Statistical Science, Southern Methodist University, Dallas, TX, USA", 
          "id": "http://www.grid.ac/institutes/grid.263864.d", 
          "name": [
            "Department of Statistical Science, Southern Methodist University, Dallas, TX, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Schucany", 
        "givenName": "William R.", 
        "id": "sg:person.01166576632.22", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01166576632.22"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1023/a:1009914915709", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1031578519", 
          "https://doi.org/10.1023/a:1009914915709"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1023/a:1012233423407", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1024818395", 
          "https://doi.org/10.1023/a:1012233423407"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2004-09", 
    "datePublishedReg": "2004-09-01", 
    "description": "In this article we implement the minimum density power divergence estimator (MDPDE) for the shape and scale parameters of the generalized Pareto distribution (GPD). The MDPDE is indexed by a constant \u03b1 \u2265 0 that controls the trade-off between robustness and efficiency. As \u03b1 increases, robustness increases and efficiency decreases. For \u03b1 = 0 the MDPDE is equivalent to the maximum likelihood estimator (MLE). We show that for \u03b1 > 0 the MDPDE for the GPD has a bounded influence function. For \u03b1 < 0.2 the MDPDE maintains good asymptotic relative efficiencies, usually above 90%. The results from a Monte Carlo study agree with these asymptotic calculations. The MDPDE is asymptotically normally distributed if the shape parameter is less than (1 + \u03b1)/(2 + \u03b1), and estimators for standard errors are readily computed under this restriction. We compare the MDPDE, MLE, Dupuis\u2019 optimally-biased robust estimator (OBRE), and Peng and Welsh\u2019s Medians estimator for the parameters. The simulations indicate that the MLE has the highest efficiency under uncontaminated GPDs. However, for the GPD contaminated with gross errors OBRE and MDPDE are more efficient than the MLE. For all the simulated models that we studied the Medians estimator had poor performance.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s10687-005-6475-6", 
    "inLanguage": "en", 
    "isAccessibleForFree": false, 
    "isPartOf": [
      {
        "id": "sg:journal.1047855", 
        "issn": [
          "1386-1999", 
          "1572-915X"
        ], 
        "name": "Extremes", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "3", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "7"
      }
    ], 
    "keywords": [
      "minimum density power divergence estimator", 
      "generalized Pareto distribution", 
      "maximum likelihood estimator", 
      "median estimator", 
      "Pareto distribution", 
      "robust estimators", 
      "divergence estimators", 
      "efficient estimation", 
      "asymptotic relative efficiency", 
      "likelihood estimator", 
      "Monte Carlo study", 
      "influence function", 
      "estimator", 
      "robustness increase", 
      "asymptotic calculations", 
      "Carlo study", 
      "shape parameters", 
      "scale parameter", 
      "standard error", 
      "Dupuis", 
      "relative efficiency", 
      "OBRE", 
      "robustness", 
      "parameters", 
      "Peng", 
      "estimation", 
      "simulations", 
      "error", 
      "efficiency", 
      "distribution", 
      "model", 
      "high efficiency", 
      "function", 
      "poor performance", 
      "performance", 
      "restriction", 
      "calculations", 
      "results", 
      "efficiency decrease", 
      "shape", 
      "article", 
      "study", 
      "increase", 
      "decrease", 
      "density power divergence estimator", 
      "power divergence estimator", 
      "good asymptotic relative efficiencies", 
      "Welsh\u2019s Medians estimator", 
      "uncontaminated GPDs", 
      "gross errors OBRE", 
      "errors OBRE"
    ], 
    "name": "Robust and Efficient Estimation for the Generalized Pareto Distribution", 
    "pagination": "237-251", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1034725432"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s10687-005-6475-6"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s10687-005-6475-6", 
      "https://app.dimensions.ai/details/publication/pub.1034725432"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2021-12-01T19:15", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20211201/entities/gbq_results/article/article_387.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s10687-005-6475-6"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s10687-005-6475-6'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s10687-005-6475-6'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10687-005-6475-6'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10687-005-6475-6'


 

This table displays all metadata directly associated to this object as RDF triples.

127 TRIPLES      22 PREDICATES      78 URIs      68 LITERALS      6 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s10687-005-6475-6 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author N1cb92a8239f84dbab040b29297af9aa9
4 schema:citation sg:pub.10.1023/a:1009914915709
5 sg:pub.10.1023/a:1012233423407
6 schema:datePublished 2004-09
7 schema:datePublishedReg 2004-09-01
8 schema:description In this article we implement the minimum density power divergence estimator (MDPDE) for the shape and scale parameters of the generalized Pareto distribution (GPD). The MDPDE is indexed by a constant α ≥ 0 that controls the trade-off between robustness and efficiency. As α increases, robustness increases and efficiency decreases. For α = 0 the MDPDE is equivalent to the maximum likelihood estimator (MLE). We show that for α > 0 the MDPDE for the GPD has a bounded influence function. For α < 0.2 the MDPDE maintains good asymptotic relative efficiencies, usually above 90%. The results from a Monte Carlo study agree with these asymptotic calculations. The MDPDE is asymptotically normally distributed if the shape parameter is less than (1 + α)/(2 + α), and estimators for standard errors are readily computed under this restriction. We compare the MDPDE, MLE, Dupuis’ optimally-biased robust estimator (OBRE), and Peng and Welsh’s Medians estimator for the parameters. The simulations indicate that the MLE has the highest efficiency under uncontaminated GPDs. However, for the GPD contaminated with gross errors OBRE and MDPDE are more efficient than the MLE. For all the simulated models that we studied the Medians estimator had poor performance.
9 schema:genre article
10 schema:inLanguage en
11 schema:isAccessibleForFree false
12 schema:isPartOf N3cb13b2d2a8a4de19b79c353688b2e07
13 Nb696bf7142d441d38fe57d3d91916074
14 sg:journal.1047855
15 schema:keywords Carlo study
16 Dupuis
17 Monte Carlo study
18 OBRE
19 Pareto distribution
20 Peng
21 Welsh’s Medians estimator
22 article
23 asymptotic calculations
24 asymptotic relative efficiency
25 calculations
26 decrease
27 density power divergence estimator
28 distribution
29 divergence estimators
30 efficiency
31 efficiency decrease
32 efficient estimation
33 error
34 errors OBRE
35 estimation
36 estimator
37 function
38 generalized Pareto distribution
39 good asymptotic relative efficiencies
40 gross errors OBRE
41 high efficiency
42 increase
43 influence function
44 likelihood estimator
45 maximum likelihood estimator
46 median estimator
47 minimum density power divergence estimator
48 model
49 parameters
50 performance
51 poor performance
52 power divergence estimator
53 relative efficiency
54 restriction
55 results
56 robust estimators
57 robustness
58 robustness increase
59 scale parameter
60 shape
61 shape parameters
62 simulations
63 standard error
64 study
65 uncontaminated GPDs
66 schema:name Robust and Efficient Estimation for the Generalized Pareto Distribution
67 schema:pagination 237-251
68 schema:productId N10d1ae3b1c7f4e9082a0eddb23045fc6
69 N30433460062845a49df74af7cd6385e8
70 schema:sameAs https://app.dimensions.ai/details/publication/pub.1034725432
71 https://doi.org/10.1007/s10687-005-6475-6
72 schema:sdDatePublished 2021-12-01T19:15
73 schema:sdLicense https://scigraph.springernature.com/explorer/license/
74 schema:sdPublisher Nced1d3a3edf84ec199e6101c6a621846
75 schema:url https://doi.org/10.1007/s10687-005-6475-6
76 sgo:license sg:explorer/license/
77 sgo:sdDataset articles
78 rdf:type schema:ScholarlyArticle
79 N10d1ae3b1c7f4e9082a0eddb23045fc6 schema:name dimensions_id
80 schema:value pub.1034725432
81 rdf:type schema:PropertyValue
82 N1cb92a8239f84dbab040b29297af9aa9 rdf:first sg:person.012227626342.30
83 rdf:rest Nf08f902916c047f580558c38d3db6694
84 N30433460062845a49df74af7cd6385e8 schema:name doi
85 schema:value 10.1007/s10687-005-6475-6
86 rdf:type schema:PropertyValue
87 N3cb13b2d2a8a4de19b79c353688b2e07 schema:issueNumber 3
88 rdf:type schema:PublicationIssue
89 Nb696bf7142d441d38fe57d3d91916074 schema:volumeNumber 7
90 rdf:type schema:PublicationVolume
91 Nced1d3a3edf84ec199e6101c6a621846 schema:name Springer Nature - SN SciGraph project
92 rdf:type schema:Organization
93 Nf08f902916c047f580558c38d3db6694 rdf:first sg:person.01166576632.22
94 rdf:rest rdf:nil
95 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
96 schema:name Mathematical Sciences
97 rdf:type schema:DefinedTerm
98 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
99 schema:name Statistics
100 rdf:type schema:DefinedTerm
101 sg:journal.1047855 schema:issn 1386-1999
102 1572-915X
103 schema:name Extremes
104 schema:publisher Springer Nature
105 rdf:type schema:Periodical
106 sg:person.01166576632.22 schema:affiliation grid-institutes:grid.263864.d
107 schema:familyName Schucany
108 schema:givenName William R.
109 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01166576632.22
110 rdf:type schema:Person
111 sg:person.012227626342.30 schema:affiliation grid-institutes:grid.42707.36
112 schema:familyName Juárez
113 schema:givenName Sergio F.
114 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012227626342.30
115 rdf:type schema:Person
116 sg:pub.10.1023/a:1009914915709 schema:sameAs https://app.dimensions.ai/details/publication/pub.1031578519
117 https://doi.org/10.1023/a:1009914915709
118 rdf:type schema:CreativeWork
119 sg:pub.10.1023/a:1012233423407 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024818395
120 https://doi.org/10.1023/a:1012233423407
121 rdf:type schema:CreativeWork
122 grid-institutes:grid.263864.d schema:alternateName Department of Statistical Science, Southern Methodist University, Dallas, TX, USA
123 schema:name Department of Statistical Science, Southern Methodist University, Dallas, TX, USA
124 rdf:type schema:Organization
125 grid-institutes:grid.42707.36 schema:alternateName Faculty of Statistics and Informatics, Veracruzana University, Xalapa, Ver, Mexico
126 schema:name Faculty of Statistics and Informatics, Veracruzana University, Xalapa, Ver, Mexico
127 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...