[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Binomial regression

From   Paul Seed <>
Subject   Re: st: Binomial regression
Date   Mon, 06 Aug 2007 13:48:09 +0100

A couple of points that have not yet been mentioned:

Binomial regression for risk differences is available for any linear model using
-binreg, rd-. This is more flexible than either -cs- or -somersd-

It would be useful if Stata extended the transformations available
for displaying results with -lincom-. At the moment, we have 2 options
(untransformed & exponential) and six labels (5 being for exponential form).
An -inverse- transformation, specifically for NNT (number needed to
treat ) would be useful for those who could use it with discretion.
An -inverse with sign change- would also be useful for NNH (number
needed to harm).

A warning would be needed that for non-significant results,
the interval can be very difficult to interpret.

Paul T Seed Msc CStat
Lecturer in Medical Statistics

King's College London
Division of Reproduction and Endocrinology

St Thomas' Hospital,
Lambeth Palace Road,
London SE1 7EH

tel (+44) (0) 20 7188 3642
fax (+44) (0) 20 7620 1227

* For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index