Comparing four bootstrap methods for stratified three-stage sampling

    Research output: Contribution to journalArticle

    4 Citations (Scopus)

    Abstract

    For a stratified three-stage sampling design with simple random sampling without replacement at each stage, only the Bernoulli bootstrap is currently available as a bootstrap for design-based inference under arbitrary sampling fractions. This article extends three other methods (the mirror-match bootstrap, the rescaling bootstrap, and the without-replacement bootstrap) to the design and conducts simulation study that estimates variances and constructs coverage intervals for a population total and selected quantiles. The without-replacement bootstrap proves the least biased of the four methods when estimating the variances of quantiles. Otherwise, the methods are comparable.

    Original languageEnglish
    Pages (from-to)193-207
    Number of pages15
    JournalJournal of Official Statistics
    Volume26
    Issue number1
    Publication statusPublished - 2010 Mar

    Fingerprint

    Bootstrap Method
    Bootstrap
    Quantile
    Design-based Inference
    Replacement
    Sampling without Replacement
    Simple Random Sampling
    Sampling Design
    Rescaling
    Bernoulli
    Biased
    Mirror
    Coverage
    Simulation Study
    Interval
    Arbitrary
    Estimate

    Keywords

    • High sampling fractions
    • Multistage sampling
    • Quantile estimation
    • Resampling methods

    ASJC Scopus subject areas

    • Statistics and Probability

    Cite this

    Comparing four bootstrap methods for stratified three-stage sampling. / Saigo, Hiroshi.

    In: Journal of Official Statistics, Vol. 26, No. 1, 03.2010, p. 193-207.

    Research output: Contribution to journalArticle

    @article{38589e60481444b99e46854a9a971ed1,
    title = "Comparing four bootstrap methods for stratified three-stage sampling",
    abstract = "For a stratified three-stage sampling design with simple random sampling without replacement at each stage, only the Bernoulli bootstrap is currently available as a bootstrap for design-based inference under arbitrary sampling fractions. This article extends three other methods (the mirror-match bootstrap, the rescaling bootstrap, and the without-replacement bootstrap) to the design and conducts simulation study that estimates variances and constructs coverage intervals for a population total and selected quantiles. The without-replacement bootstrap proves the least biased of the four methods when estimating the variances of quantiles. Otherwise, the methods are comparable.",
    keywords = "High sampling fractions, Multistage sampling, Quantile estimation, Resampling methods",
    author = "Hiroshi Saigo",
    year = "2010",
    month = "3",
    language = "English",
    volume = "26",
    pages = "193--207",
    journal = "Journal of Official Statistics",
    issn = "0282-423X",
    publisher = "Statistics Sweden",
    number = "1",

    }

    TY - JOUR

    T1 - Comparing four bootstrap methods for stratified three-stage sampling

    AU - Saigo, Hiroshi

    PY - 2010/3

    Y1 - 2010/3

    N2 - For a stratified three-stage sampling design with simple random sampling without replacement at each stage, only the Bernoulli bootstrap is currently available as a bootstrap for design-based inference under arbitrary sampling fractions. This article extends three other methods (the mirror-match bootstrap, the rescaling bootstrap, and the without-replacement bootstrap) to the design and conducts simulation study that estimates variances and constructs coverage intervals for a population total and selected quantiles. The without-replacement bootstrap proves the least biased of the four methods when estimating the variances of quantiles. Otherwise, the methods are comparable.

    AB - For a stratified three-stage sampling design with simple random sampling without replacement at each stage, only the Bernoulli bootstrap is currently available as a bootstrap for design-based inference under arbitrary sampling fractions. This article extends three other methods (the mirror-match bootstrap, the rescaling bootstrap, and the without-replacement bootstrap) to the design and conducts simulation study that estimates variances and constructs coverage intervals for a population total and selected quantiles. The without-replacement bootstrap proves the least biased of the four methods when estimating the variances of quantiles. Otherwise, the methods are comparable.

    KW - High sampling fractions

    KW - Multistage sampling

    KW - Quantile estimation

    KW - Resampling methods

    UR - http://www.scopus.com/inward/record.url?scp=77951778451&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=77951778451&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:77951778451

    VL - 26

    SP - 193

    EP - 207

    JO - Journal of Official Statistics

    JF - Journal of Official Statistics

    SN - 0282-423X

    IS - 1

    ER -