WILLIAMS: Stata module to estimate logistic regression via Williams procedure

Statistical Software Components from Boston College Department of Economics

Abstract: The Williams procedure, originally written in GLIM, helps accomodate for overdispersion in binomial (proportional) models. It is not a post facto sort of adjustment; rather there is an adjustment made to the variance at each iteration in the IRLS routine. The Williams procedure iteratively reduces the Chi2-based dispersion to approximately 1.0. An extra parameter, called phi, is used to scale the variance function. It changes with each iteration. The value of phi which results in a chi2-based dispersion of 1.0 is then used to weight a standard grouped logistic regression. The final weighted logistic regression displays p-values based on t-statistics per recommendation of D. Collett (1991), Modeling Binary Data, Chapman & Hall, p 196.

Language: Stata
Date: Written 1998-09-10

Downloads: (external link)
http://fmwww.bc.edu/repec/bocode/w/williams.ado program code (text/plain)
http://fmwww.bc.edu/repec/bocode/_/_binom.ado program code (text/plain)
http://fmwww.bc.edu/repec/bocode/w/williams.hlp help file (text/plain)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Ordering information: This software item can be ordered from
http://repec.org/docs/ssc.php

Access Statistics for this software item

More software in Statistical Software Components from Boston College Department of Economics
Address: Boston College, 140 Commonwealth Avenue, Chestnut Hill MA 02467 USA
Contact information at EDIRC.
Series data maintained by Christopher F Baum ().