Scribd Logo
Upload

Welcome to Scribd!

  • Software For Computing Statistical Intervals: Table 1

    Uploaded by

    Edwin Johny Asnate Salazar
    17 views2 pages
    The document summarizes a spreadsheet that lists 67 types of statistical intervals that are useful in applications and indicates what software can be used to calculate each interval. The spreadsheet includes the statistical method used to construct each interval and a reference to the section in a book by Meeker, Hahn, and Escobar (2017) that discusses the interval and calculation. Tables in the document explain codes used in the spreadsheet to identify the statistical method and type of data for which each interval applies. The information on software capabilities was obtained from vendors but not verified. The authors plan to update the spreadsheet as new software developments occur.

    Original Description:

    Original Title

    StatisticalSoftwareCapabilities

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    The document summarizes a spreadsheet that lists 67 types of statistical intervals that are useful in applications and indicates what software can be used to calculate each interval. The spreadsheet includes the statistical method used to construct each interval and a reference to the section in a book by Meeker, Hahn, and Escobar (2017) that discusses the interval and calculation. Tables in the document explain codes used in the spreadsheet to identify the statistical method and type of data for which each interval applies. The information on software capabilities was obtained from vendors but not verified. The authors plan to update the spreadsheet as new software developments occur.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as docx, pdf, or txt
    17 views2 pages

    Software For Computing Statistical Intervals: Table 1

    Uploaded by

    Edwin Johny Asnate Salazar
    The document summarizes a spreadsheet that lists 67 types of statistical intervals that are useful in applications and indicates what software can be used to calculate each interval. The spreadsheet includes the statistical method used to construct each interval and a reference to the section in a book by Meeker, Hahn, and Escobar (2017) that discusses the interval and calculation. Tables in the document explain codes used in the spreadsheet to identify the statistical method and type of data for which each interval applies. The information on software capabilities was obtained from vendors but not verified. The authors plan to update the spreadsheet as new software developments occur.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as docx, pdf, or txt
    of 2

    Software for Computing Statistical Intervals

    The accompanying spreadsheet enumerates the capabilities of generally (or commercially) available
    software packages for computing a wide variety of confidence, tolerance, and prediction intervals
    required in practical applications and statistical studies. All of these statistical intervals are described, in
    varying levels of detail, in the second edition of Statistical Intervals: A Guide for Practitioners and
    Researchers by Meeker, Hahn, and Escobar (2017) (MHE2017).

    Exact statistical intervals exist for models using an underlying normal distribution and for a few other
    special cases. Otherwise, approximate (and in some cases conservative) statistical interval procedures
    must be used. As described in MHE2017, many different methods are now available for computing
    needed statistical intervals.

    The attached spreadsheet lists 67 different kinds of statistical intervals that are useful or required in
    practical applications (e.g., confidence interval for a normal distribution quantile, distribution free
    control-the-center tolerance interval, Weibull distribution prediction interval to contain k of m future
    observations) for different types of data (e.g., complete, interval censored) and indicates software that
    calculates such intervals. The tabulation also shows the statistical interval method used and the section
    in MHE2017 where the interval and its calculation are discussed.

    Table 1 explains the letter code used in the spreadsheet to identify the statistical method used in
    MHE2017 to constructing each of the intervals. Table 2 explains the letter code used in the spreadsheet
    for describing the type of data for which the stated interval applies.

    Table 1
    Statistical Interval Method Codes
    B Bayesian with user-specified prior distribution
    BS Parametric bootstrap or simulation (exact for location-scale and complete data or type 2
    censoring, approximate otherwise) or a generalized pivotal quantity
    C Conservative
    E Exact (coverage probability is exactly equal to nominal confidence level) based on a pivotal
    quantity
    IOS Interpolated order statistics (approximately distribution free; see OS)
    J Bayesian with a Jeffreys prior
    L Approximate, based on inverting a likelihood ratio test
    N Normal distribution approximation
    NPBS Nonparametric bootstrap
    OS Based on selected order statistics (distribution-free)
    OA Other approximate method(s)
    S Approximate, based on inverting a score test
    W Approximate, based on a Wald statistic approximation to the likelihood method
    Table 2
    Data Type Codes
    A Arbitrarily censored (mixture of left, right, interval, over-lapping intervals)
    C Complete (no censoring)
    I Interval-censored
    L Left-censored
    R Right-censored

    The information on available software in the spreadsheet was obtained directly from the software
    vendor/developer and was not verified by us. The matrix indicates which capabilities are available in the
    different software offerings but not the specific routine, etc. within the software. Footnote pages for
    individual software products provide such additional information. Please contact the vendor/developer
    directly if you have any questions about a particular software package or feature.

    We plan to update the Excel workbook as new software capabilities are released and we are informed
    about them.

    William Q. Meeker
    Gerald J. Hahn
    Luis A. Escobar
    March 18, 2017

    You might also like