Bayes factor asymptotics for variable selection in the Gaussian process framework View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2021-09-20

AUTHORS

Minerva Mukhopadhyay, Sourabh Bhattacharya

ABSTRACT

We investigate Bayesian variable selection in models driven by Gaussian processes, which allows us to treat linear, nonlinear and nonparametric models, in conjunction with even dependent setups, in the same vein. We consider the Bayes factor route to variable selection, and develop a general asymptotic theory for the Gaussian process framework in the “large p, large n” settings even with p≫n\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$p\gg n$$\end{document}, establishing almost sure exponential convergence of the Bayes factor under appropriately mild conditions. The fixed p setup is included as a special case. To illustrate, we apply our result to variable selection in linear regression, Gaussian process model with squared exponential covariance function accommodating the covariates, and a first-order autoregressive process with time-varying covariates. We also follow up our theoretical investigations with ample simulation experiments in the above regression contexts and variable selection in a real, riboflavin data consisting of 71 observations and 4088 covariates. For implementation of variable selection using Bayes factors, we develop a novel and effective general-purpose transdimensional, transformation-based Markov chain Monte Carlo algorithm, which has played a crucial role in simulated and real data applications. More... »

PAGES

1-33

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s10463-021-00810-6

DOI

http://dx.doi.org/10.1007/s10463-021-00810-6

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1141239613


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": "Department of Mathematics and Statistics, Indian Institute of Technology, 208016, Kanpur, India", 
          "id": "http://www.grid.ac/institutes/grid.417965.8", 
          "name": [
            "Department of Mathematics and Statistics, Indian Institute of Technology, 208016, Kanpur, India"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Mukhopadhyay", 
        "givenName": "Minerva", 
        "id": "sg:person.014622734305.51", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014622734305.51"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Interdisciplinary Statistical Research Unit, Indian Statistical Institute, 203 B.T. Road, 700108, Kolkata, India", 
          "id": "http://www.grid.ac/institutes/grid.39953.35", 
          "name": [
            "Interdisciplinary Statistical Research Unit, Indian Statistical Institute, 203 B.T. Road, 700108, Kolkata, India"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Bhattacharya", 
        "givenName": "Sourabh", 
        "id": "sg:person.01012545423.95", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01012545423.95"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/s10463-014-0483-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1052704086", 
          "https://doi.org/10.1007/s10463-014-0483-8"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-5945-4", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1019309908", 
          "https://doi.org/10.1007/978-1-4419-5945-4"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11749-006-0039-1", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021521046", 
          "https://doi.org/10.1007/s11749-006-0039-1"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2021-09-20", 
    "datePublishedReg": "2021-09-20", 
    "description": "We investigate Bayesian variable selection in models driven by Gaussian processes, which allows us to treat linear, nonlinear and nonparametric models, in conjunction with even dependent setups, in the same vein. We consider the Bayes factor route to variable selection, and develop a general asymptotic theory for the Gaussian process framework in the \u201clarge p, large n\u201d settings even with p\u226bn\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym}\n\t\t\t\t\\usepackage{amsfonts}\n\t\t\t\t\\usepackage{amssymb}\n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$p\\gg n$$\\end{document}, establishing almost sure exponential convergence of the Bayes factor under appropriately mild conditions. The fixed p setup is included as a special case. To illustrate, we apply our result to variable selection in linear regression, Gaussian process model with squared exponential covariance function accommodating the covariates, and a first-order autoregressive process with time-varying covariates. We also follow up our theoretical investigations with ample simulation experiments in the above regression contexts and variable selection in a real, riboflavin data consisting of 71 observations and 4088 covariates. For implementation of variable selection using Bayes factors, we develop a novel and effective general-purpose transdimensional, transformation-based Markov chain Monte Carlo algorithm, which has played a crucial role in simulated and real data applications.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s10463-021-00810-6", 
    "inLanguage": "en", 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1041657", 
        "issn": [
          "0020-3157", 
          "1572-9052"
        ], 
        "name": "Annals of the Institute of Statistical Mathematics", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }
    ], 
    "keywords": [
      "Gaussian process framework", 
      "variable selection", 
      "Markov chain Monte Carlo algorithm", 
      "process framework", 
      "Bayes factors", 
      "general asymptotic theory", 
      "sure exponential convergence", 
      "squared exponential covariance function", 
      "Bayesian variable selection", 
      "Gaussian process model", 
      "exponential covariance function", 
      "Monte Carlo algorithm", 
      "real data application", 
      "data applications", 
      "asymptotic theory", 
      "first-order autoregressive process", 
      "exponential convergence", 
      "Carlo algorithm", 
      "covariance function", 
      "Gaussian process", 
      "dependent setups", 
      "simulation experiments", 
      "regression context", 
      "process model", 
      "nonparametric model", 
      "autoregressive process", 
      "special case", 
      "theoretical investigation", 
      "framework", 
      "algorithm", 
      "asymptotics", 
      "selection", 
      "time-varying covariates", 
      "convergence", 
      "model", 
      "implementation", 
      "linear regression", 
      "setup", 
      "theory", 
      "applications", 
      "covariates", 
      "same vein", 
      "mild conditions", 
      "process", 
      "context", 
      "data", 
      "function", 
      "experiments", 
      "cases", 
      "observations", 
      "conjunction", 
      "crucial role", 
      "conditions", 
      "results", 
      "regression", 
      "setting", 
      "route", 
      "investigation", 
      "factors", 
      "role", 
      "vein", 
      "Bayes factor route", 
      "factor route", 
      "ample simulation experiments", 
      "above regression contexts", 
      "riboflavin data", 
      "transformation-based Markov chain Monte Carlo algorithm", 
      "chain Monte Carlo algorithm", 
      "Bayes factor asymptotics", 
      "factor asymptotics"
    ], 
    "name": "Bayes factor asymptotics for variable selection in the Gaussian process framework", 
    "pagination": "1-33", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1141239613"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s10463-021-00810-6"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s10463-021-00810-6", 
      "https://app.dimensions.ai/details/publication/pub.1141239613"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-01-01T18:58", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/article/article_890.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s10463-021-00810-6"
  }
]
 

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/s10463-021-00810-6'

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/s10463-021-00810-6'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10463-021-00810-6'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10463-021-00810-6'


 

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

144 TRIPLES      22 PREDICATES      96 URIs      85 LITERALS      4 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s10463-021-00810-6 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author N352e611a8bb54e1ab2ee35547cc1bdec
4 schema:citation sg:pub.10.1007/978-1-4419-5945-4
5 sg:pub.10.1007/s10463-014-0483-8
6 sg:pub.10.1007/s11749-006-0039-1
7 schema:datePublished 2021-09-20
8 schema:datePublishedReg 2021-09-20
9 schema:description We investigate Bayesian variable selection in models driven by Gaussian processes, which allows us to treat linear, nonlinear and nonparametric models, in conjunction with even dependent setups, in the same vein. We consider the Bayes factor route to variable selection, and develop a general asymptotic theory for the Gaussian process framework in the “large p, large n” settings even with p≫n\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$p\gg n$$\end{document}, establishing almost sure exponential convergence of the Bayes factor under appropriately mild conditions. The fixed p setup is included as a special case. To illustrate, we apply our result to variable selection in linear regression, Gaussian process model with squared exponential covariance function accommodating the covariates, and a first-order autoregressive process with time-varying covariates. We also follow up our theoretical investigations with ample simulation experiments in the above regression contexts and variable selection in a real, riboflavin data consisting of 71 observations and 4088 covariates. For implementation of variable selection using Bayes factors, we develop a novel and effective general-purpose transdimensional, transformation-based Markov chain Monte Carlo algorithm, which has played a crucial role in simulated and real data applications.
10 schema:genre article
11 schema:inLanguage en
12 schema:isAccessibleForFree true
13 schema:isPartOf sg:journal.1041657
14 schema:keywords Bayes factor asymptotics
15 Bayes factor route
16 Bayes factors
17 Bayesian variable selection
18 Carlo algorithm
19 Gaussian process
20 Gaussian process framework
21 Gaussian process model
22 Markov chain Monte Carlo algorithm
23 Monte Carlo algorithm
24 above regression contexts
25 algorithm
26 ample simulation experiments
27 applications
28 asymptotic theory
29 asymptotics
30 autoregressive process
31 cases
32 chain Monte Carlo algorithm
33 conditions
34 conjunction
35 context
36 convergence
37 covariance function
38 covariates
39 crucial role
40 data
41 data applications
42 dependent setups
43 experiments
44 exponential convergence
45 exponential covariance function
46 factor asymptotics
47 factor route
48 factors
49 first-order autoregressive process
50 framework
51 function
52 general asymptotic theory
53 implementation
54 investigation
55 linear regression
56 mild conditions
57 model
58 nonparametric model
59 observations
60 process
61 process framework
62 process model
63 real data application
64 regression
65 regression context
66 results
67 riboflavin data
68 role
69 route
70 same vein
71 selection
72 setting
73 setup
74 simulation experiments
75 special case
76 squared exponential covariance function
77 sure exponential convergence
78 theoretical investigation
79 theory
80 time-varying covariates
81 transformation-based Markov chain Monte Carlo algorithm
82 variable selection
83 vein
84 schema:name Bayes factor asymptotics for variable selection in the Gaussian process framework
85 schema:pagination 1-33
86 schema:productId N354de227a6b6405baff48a0e3e7a5197
87 N651c033055b146c393f66e7a039905e6
88 schema:sameAs https://app.dimensions.ai/details/publication/pub.1141239613
89 https://doi.org/10.1007/s10463-021-00810-6
90 schema:sdDatePublished 2022-01-01T18:58
91 schema:sdLicense https://scigraph.springernature.com/explorer/license/
92 schema:sdPublisher N706e82321abb44c38bcdec10fc35c801
93 schema:url https://doi.org/10.1007/s10463-021-00810-6
94 sgo:license sg:explorer/license/
95 sgo:sdDataset articles
96 rdf:type schema:ScholarlyArticle
97 N352e611a8bb54e1ab2ee35547cc1bdec rdf:first sg:person.014622734305.51
98 rdf:rest N5d98a299ef63462c9a60146334e533ec
99 N354de227a6b6405baff48a0e3e7a5197 schema:name doi
100 schema:value 10.1007/s10463-021-00810-6
101 rdf:type schema:PropertyValue
102 N5d98a299ef63462c9a60146334e533ec rdf:first sg:person.01012545423.95
103 rdf:rest rdf:nil
104 N651c033055b146c393f66e7a039905e6 schema:name dimensions_id
105 schema:value pub.1141239613
106 rdf:type schema:PropertyValue
107 N706e82321abb44c38bcdec10fc35c801 schema:name Springer Nature - SN SciGraph project
108 rdf:type schema:Organization
109 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
110 schema:name Mathematical Sciences
111 rdf:type schema:DefinedTerm
112 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
113 schema:name Statistics
114 rdf:type schema:DefinedTerm
115 sg:journal.1041657 schema:issn 0020-3157
116 1572-9052
117 schema:name Annals of the Institute of Statistical Mathematics
118 schema:publisher Springer Nature
119 rdf:type schema:Periodical
120 sg:person.01012545423.95 schema:affiliation grid-institutes:grid.39953.35
121 schema:familyName Bhattacharya
122 schema:givenName Sourabh
123 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01012545423.95
124 rdf:type schema:Person
125 sg:person.014622734305.51 schema:affiliation grid-institutes:grid.417965.8
126 schema:familyName Mukhopadhyay
127 schema:givenName Minerva
128 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014622734305.51
129 rdf:type schema:Person
130 sg:pub.10.1007/978-1-4419-5945-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1019309908
131 https://doi.org/10.1007/978-1-4419-5945-4
132 rdf:type schema:CreativeWork
133 sg:pub.10.1007/s10463-014-0483-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052704086
134 https://doi.org/10.1007/s10463-014-0483-8
135 rdf:type schema:CreativeWork
136 sg:pub.10.1007/s11749-006-0039-1 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021521046
137 https://doi.org/10.1007/s11749-006-0039-1
138 rdf:type schema:CreativeWork
139 grid-institutes:grid.39953.35 schema:alternateName Interdisciplinary Statistical Research Unit, Indian Statistical Institute, 203 B.T. Road, 700108, Kolkata, India
140 schema:name Interdisciplinary Statistical Research Unit, Indian Statistical Institute, 203 B.T. Road, 700108, Kolkata, India
141 rdf:type schema:Organization
142 grid-institutes:grid.417965.8 schema:alternateName Department of Mathematics and Statistics, Indian Institute of Technology, 208016, Kanpur, India
143 schema:name Department of Mathematics and Statistics, Indian Institute of Technology, 208016, Kanpur, India
144 rdf:type schema:Organization
 




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


...