Bayesian Analysis of Nonnegative Data Using Dependency-Extended Two-Part Models View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2021-08-25

AUTHORS

Mariana Rodrigues-Motta, Johannes Forkman

ABSTRACT

This article is motivated by the challenge of analysing an agricultural field experiment with observations that are positive on a continuous scale or zero. Such data can be analysed using two-part models, where the distribution is a mixture of a positive distribution and a Bernoulli distribution. However, traditional two-part models do not include any dependencies between the two parts of the model. Since the probability of zero is anticipated to be high when the expected value of the positive part is low, and the other way around, this article introduces dependency-extended two-part models. In addition, these extensions allow for modelling the median instead of the mean, which has advantages when distributions are skewed. The motivating example is an incomplete block trial comparing ten treatments against weed. Gamma and lognormal distributions were used for the positive response, although any density on the support of real numbers can be accommodated. In a cross-validation study, the proposed new models were compared with each other and with a baseline model without dependencies. Model performance and sensitivity to choice of priors were investigated through simulation. A dependency-extended two-part model for the median of the lognormal distribution performed best with regard to mean square error in prediction. Supplementary materials accompanying this paper appear online. More... »

PAGES

201-221

References to SciGraph publications

  • 1988. Transformation and Weighting in Regression in NONE
  • 2002-09. A bayesian approach to regional and local-area prediction from crop variety trials in JOURNAL OF AGRICULTURAL, BIOLOGICAL AND ENVIRONMENTAL STATISTICS
  • 2020-05-05. History of the Statistical Design of Agricultural Experiments in JOURNAL OF AGRICULTURAL, BIOLOGICAL AND ENVIRONMENTAL STATISTICS
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s13253-021-00467-x

    DOI

    http://dx.doi.org/10.1007/s13253-021-00467-x

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1140638622


    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 Statistics, State University of Campinas, S\u00e3o Paulo, Brazil", 
              "id": "http://www.grid.ac/institutes/grid.411087.b", 
              "name": [
                "Department of Statistics, State University of Campinas, S\u00e3o Paulo, Brazil"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Rodrigues-Motta", 
            "givenName": "Mariana", 
            "id": "sg:person.01100716043.72", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01100716043.72"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Crop Production Ecology, Swedish University of Agricultural Sciences, PO Box 7043, 750 07, Uppsala, Sweden", 
              "id": "http://www.grid.ac/institutes/grid.6341.0", 
              "name": [
                "Department of Crop Production Ecology, Swedish University of Agricultural Sciences, PO Box 7043, 750 07, Uppsala, Sweden"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Forkman", 
            "givenName": "Johannes", 
            "id": "sg:person.01256742405.14", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01256742405.14"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/s13253-020-00394-3", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1127397892", 
              "https://doi.org/10.1007/s13253-020-00394-3"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1198/108571102230", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1040861262", 
              "https://doi.org/10.1198/108571102230"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-1-4899-2873-3", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1109705816", 
              "https://doi.org/10.1007/978-1-4899-2873-3"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2021-08-25", 
        "datePublishedReg": "2021-08-25", 
        "description": "This article is motivated by the challenge of analysing an agricultural field experiment with observations that are positive on a continuous scale or zero. Such data can be analysed using two-part models, where the distribution is a mixture of a positive distribution and a Bernoulli distribution. However, traditional two-part models do not include any dependencies between the two parts of the model. Since the probability of zero is anticipated to be high when the expected value of the positive part is low, and the other way around, this article introduces dependency-extended two-part models. In addition, these extensions allow for modelling the median instead of the mean, which has advantages when distributions are skewed. The motivating example is an incomplete block trial comparing ten treatments against weed. Gamma and lognormal distributions were used for the positive response, although any density on the support of real numbers can be accommodated. In a cross-validation study, the proposed new models were compared with each other and with a baseline model without dependencies. Model performance and sensitivity to choice of priors were investigated through simulation. A dependency-extended two-part model for the median of the lognormal distribution performed best with regard to mean square error in prediction. Supplementary materials accompanying this paper appear online.", 
        "genre": "article", 
        "id": "sg:pub.10.1007/s13253-021-00467-x", 
        "inLanguage": "en", 
        "isAccessibleForFree": true, 
        "isPartOf": [
          {
            "id": "sg:journal.1134206", 
            "issn": [
              "1085-7117", 
              "1537-2693"
            ], 
            "name": "Journal of Agricultural, Biological and Environmental Statistics", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "2", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "27"
          }
        ], 
        "keywords": [
          "choice of priors", 
          "lognormal distribution", 
          "Bernoulli distribution", 
          "two-part model", 
          "positive distribution", 
          "real numbers", 
          "Bayesian analysis", 
          "positive part", 
          "agricultural field experiments", 
          "square error", 
          "supplementary material", 
          "new model", 
          "model performance", 
          "model", 
          "distribution", 
          "priors", 
          "cross-validation study", 
          "such data", 
          "baseline model", 
          "simulations", 
          "probability", 
          "continuous scale", 
          "error", 
          "extension", 
          "dependency", 
          "prediction", 
          "density", 
          "number", 
          "means", 
          "performance", 
          "advantages", 
          "observations", 
          "experiments", 
          "choice", 
          "gamma", 
          "values", 
          "part", 
          "way", 
          "article", 
          "analysis", 
          "scale", 
          "data", 
          "materials", 
          "mixture", 
          "addition", 
          "sensitivity", 
          "challenges", 
          "median", 
          "regard", 
          "study", 
          "field experiment", 
          "response", 
          "support", 
          "positive response", 
          "block trial", 
          "example", 
          "treatment", 
          "trials", 
          "weeds", 
          "paper"
        ], 
        "name": "Bayesian Analysis of Nonnegative Data Using Dependency-Extended Two-Part Models", 
        "pagination": "201-221", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1140638622"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s13253-021-00467-x"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s13253-021-00467-x", 
          "https://app.dimensions.ai/details/publication/pub.1140638622"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-05-20T07:39", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220519/entities/gbq_results/article/article_903.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1007/s13253-021-00467-x"
      }
    ]
     

    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/s13253-021-00467-x'

    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/s13253-021-00467-x'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s13253-021-00467-x'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s13253-021-00467-x'


     

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

    140 TRIPLES      22 PREDICATES      87 URIs      76 LITERALS      6 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s13253-021-00467-x schema:about anzsrc-for:01
    2 anzsrc-for:0104
    3 schema:author N0e12480534df41c0b0d62f4fcf4cff98
    4 schema:citation sg:pub.10.1007/978-1-4899-2873-3
    5 sg:pub.10.1007/s13253-020-00394-3
    6 sg:pub.10.1198/108571102230
    7 schema:datePublished 2021-08-25
    8 schema:datePublishedReg 2021-08-25
    9 schema:description This article is motivated by the challenge of analysing an agricultural field experiment with observations that are positive on a continuous scale or zero. Such data can be analysed using two-part models, where the distribution is a mixture of a positive distribution and a Bernoulli distribution. However, traditional two-part models do not include any dependencies between the two parts of the model. Since the probability of zero is anticipated to be high when the expected value of the positive part is low, and the other way around, this article introduces dependency-extended two-part models. In addition, these extensions allow for modelling the median instead of the mean, which has advantages when distributions are skewed. The motivating example is an incomplete block trial comparing ten treatments against weed. Gamma and lognormal distributions were used for the positive response, although any density on the support of real numbers can be accommodated. In a cross-validation study, the proposed new models were compared with each other and with a baseline model without dependencies. Model performance and sensitivity to choice of priors were investigated through simulation. A dependency-extended two-part model for the median of the lognormal distribution performed best with regard to mean square error in prediction. Supplementary materials accompanying this paper appear online.
    10 schema:genre article
    11 schema:inLanguage en
    12 schema:isAccessibleForFree true
    13 schema:isPartOf N9d6e1e0248e74a84ace704d695093656
    14 Ncf1294f273b5462a806e8ae4615e1fdf
    15 sg:journal.1134206
    16 schema:keywords Bayesian analysis
    17 Bernoulli distribution
    18 addition
    19 advantages
    20 agricultural field experiments
    21 analysis
    22 article
    23 baseline model
    24 block trial
    25 challenges
    26 choice
    27 choice of priors
    28 continuous scale
    29 cross-validation study
    30 data
    31 density
    32 dependency
    33 distribution
    34 error
    35 example
    36 experiments
    37 extension
    38 field experiment
    39 gamma
    40 lognormal distribution
    41 materials
    42 means
    43 median
    44 mixture
    45 model
    46 model performance
    47 new model
    48 number
    49 observations
    50 paper
    51 part
    52 performance
    53 positive distribution
    54 positive part
    55 positive response
    56 prediction
    57 priors
    58 probability
    59 real numbers
    60 regard
    61 response
    62 scale
    63 sensitivity
    64 simulations
    65 square error
    66 study
    67 such data
    68 supplementary material
    69 support
    70 treatment
    71 trials
    72 two-part model
    73 values
    74 way
    75 weeds
    76 schema:name Bayesian Analysis of Nonnegative Data Using Dependency-Extended Two-Part Models
    77 schema:pagination 201-221
    78 schema:productId N1d1da1c0b8c7413aa2068b967139895c
    79 N99024d0d2ea04fadb09282dcdf8ce87a
    80 schema:sameAs https://app.dimensions.ai/details/publication/pub.1140638622
    81 https://doi.org/10.1007/s13253-021-00467-x
    82 schema:sdDatePublished 2022-05-20T07:39
    83 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    84 schema:sdPublisher N71c3dd800dae46af888625679923a8ee
    85 schema:url https://doi.org/10.1007/s13253-021-00467-x
    86 sgo:license sg:explorer/license/
    87 sgo:sdDataset articles
    88 rdf:type schema:ScholarlyArticle
    89 N0e12480534df41c0b0d62f4fcf4cff98 rdf:first sg:person.01100716043.72
    90 rdf:rest N3cd1d704ece54ce68e0eaf258aa80980
    91 N1d1da1c0b8c7413aa2068b967139895c schema:name doi
    92 schema:value 10.1007/s13253-021-00467-x
    93 rdf:type schema:PropertyValue
    94 N3cd1d704ece54ce68e0eaf258aa80980 rdf:first sg:person.01256742405.14
    95 rdf:rest rdf:nil
    96 N71c3dd800dae46af888625679923a8ee schema:name Springer Nature - SN SciGraph project
    97 rdf:type schema:Organization
    98 N99024d0d2ea04fadb09282dcdf8ce87a schema:name dimensions_id
    99 schema:value pub.1140638622
    100 rdf:type schema:PropertyValue
    101 N9d6e1e0248e74a84ace704d695093656 schema:volumeNumber 27
    102 rdf:type schema:PublicationVolume
    103 Ncf1294f273b5462a806e8ae4615e1fdf schema:issueNumber 2
    104 rdf:type schema:PublicationIssue
    105 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
    106 schema:name Mathematical Sciences
    107 rdf:type schema:DefinedTerm
    108 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
    109 schema:name Statistics
    110 rdf:type schema:DefinedTerm
    111 sg:journal.1134206 schema:issn 1085-7117
    112 1537-2693
    113 schema:name Journal of Agricultural, Biological and Environmental Statistics
    114 schema:publisher Springer Nature
    115 rdf:type schema:Periodical
    116 sg:person.01100716043.72 schema:affiliation grid-institutes:grid.411087.b
    117 schema:familyName Rodrigues-Motta
    118 schema:givenName Mariana
    119 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01100716043.72
    120 rdf:type schema:Person
    121 sg:person.01256742405.14 schema:affiliation grid-institutes:grid.6341.0
    122 schema:familyName Forkman
    123 schema:givenName Johannes
    124 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01256742405.14
    125 rdf:type schema:Person
    126 sg:pub.10.1007/978-1-4899-2873-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1109705816
    127 https://doi.org/10.1007/978-1-4899-2873-3
    128 rdf:type schema:CreativeWork
    129 sg:pub.10.1007/s13253-020-00394-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1127397892
    130 https://doi.org/10.1007/s13253-020-00394-3
    131 rdf:type schema:CreativeWork
    132 sg:pub.10.1198/108571102230 schema:sameAs https://app.dimensions.ai/details/publication/pub.1040861262
    133 https://doi.org/10.1198/108571102230
    134 rdf:type schema:CreativeWork
    135 grid-institutes:grid.411087.b schema:alternateName Department of Statistics, State University of Campinas, São Paulo, Brazil
    136 schema:name Department of Statistics, State University of Campinas, São Paulo, Brazil
    137 rdf:type schema:Organization
    138 grid-institutes:grid.6341.0 schema:alternateName Department of Crop Production Ecology, Swedish University of Agricultural Sciences, PO Box 7043, 750 07, Uppsala, Sweden
    139 schema:name Department of Crop Production Ecology, Swedish University of Agricultural Sciences, PO Box 7043, 750 07, Uppsala, Sweden
    140 rdf:type schema:Organization
     




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


    ...