Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • New version of esetran on SSC

    Thanks as always to Kit Baum, a new version of the esetran package is now available for download from SSC. In Stata, use the ssc command to do this, or adoupdate if you already have an old version of esetran.

    The esetran package is described as below on my website. The new version adds 2 additional transformations for estimates of proportions, namely the decreasing log-log transformation and the increasing complementary log-log transformation.

    Best wishes

    Roger

    ---------------------------------------------------------------------------------
    package esetran from http://www.rogernewsonresources.org.uk/stata10
    ---------------------------------------------------------------------------------

    TITLE
    esetran: Transforming estimates and standard errors in parmest resultssets

    DESCRIPTION/AUTHOR(S)
    esetran is designed for use in parmest resultssets, which have one
    observation per estimated parameter and data on parameter estimates.
    It inputs 2 user-specified variables, containing the estimates and the
    standard errors, and replaces them with the estimates and standard
    errors of the same parameters after a user-specified transformation,
    promoting their storage types to double precision if necessary.
    Parameter values at the boundaries of the parameter range (such as the
    logit of 1 or 0 or the hyperbolic arctangent of 1 or -1) are set to
    sensible non-missing boundary values. The user can then use the
    parmcip option of the parmest package to recalculate the t- and
    z-statistics, symmetric confidence limits and P-values for the
    transformed parameters, and use endpoint transformations on the
    estimates and confidence limits to produce asymmetric confidence
    intervals for back-transformed parameters. This practice is
    especially useful if the user has produced symmetric confidence
    intervals for scenario proportions and their differences, using
    margins after a logistic regression, and needs to replace them with
    asymmetric confidence intervals, which are more likely to have the
    correct coverage probability.

    Author: Roger Newson
    Distribution-Date: 31july2015
    Stata-Version: 10

    INSTALLATION FILES (click here to install)
    esetran.ado
    esetran.sthlp
    ---------------------------------------------------------------------------------
    (click here to return to the previous screen)

Working...
X