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

Re: st: Help please re Ocratio

From   "Jon Heron (ALSPAC)" <>
Subject   Re: st: Help please re Ocratio
Date   Wed, 31 Oct 2007 12:56:14 +0000

I could use an extra bit of clarification:

It sounds like, if I want to fit an ocratio model and then examine the
individual pairs of effects I either fit two logistics

1 [=1] versus 2 & 3[=0]
2 [=1] versus 3 [=0]

and compare with an ocratio on an outcome ordered 1,2,3

or fit two logits

1 [=0] versus 2 & 3[=1]
2 [=0] versus 3 [=1]

and compare with an ocratio with outcome ordered the other way i.e. 3,2,1

If this is the case, and a quick experiment with Stata suggests it is,
and since the continuation ratio model is not reversible (unlike
proportional odds), how do I decide which one to do?

I guess the former

Dr Jon Heron
Statistics Team Leader
ALSPAC, Dept of Social Medicine
24 Tyndall Avenue
Bristol BS8 1TQ
Tel: 0117 3311616
Fax: 0117 3311704

--On 30 October 2007 08:50 -0400 Anders Alexandersson <> wrote:

Please provide complete references. Rory Wolfe's command -ocratio-
uses a "forward stopping" continuation ratio model, P[Y=j|Y>=j] (i.e.,
like "Armstrong and Sloan" below), but with reverse sign of the slope
compared with logistic regression on expanded data so that the sign of
the slope is the same as in the usual proportional odds model.

Anders Alexandersson

Jon Heron (ALSPAC) <> wrote:
 I would like to fit a continuation ratio model to a 6-level outcome
 using ocratio. I would then like to assess the constancy of model
 parameters by fitting a series of logistic models

 A quick lit search is showing variability in which comparisons are being
 made with a CR model.

  e.g. for 3-level outcome - 1/2/3

  Armstrong and Sloan (1989)
    1 versus 2 & 3
    2 versus 3

  Manor, Matthews and Power (2000)
    1 versus 2
    1 & 2 versus 3

  Greenland (1994)
    2 & 3 versus 1,2 & 3
    3 versus 2 & 3

 I imagine these are all permitted alternatives, but i would like to know
 which particular comparisons are made using ocratio so I can fit the
 appropriate logistic models.
*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index