halvorsen kjetilh at umsanet.edu.bo
Tue Jul 25 00:29:42 CEST 2000

```I have experimented with the cheese data example from McCullagh&Nelder,
page 175. With a proportional odds model they obtain a residual deviance
of
20.31.

Estimating the same model with polr(MASS) gives a residual deviance of
762.11 !,  while using ordglm(gnlm) gives a deviance of 523.94.  Can
anybody explain these differences?

The data frame with the data are:

> cheese
Cheese  N Response nResponse
1       A  1      III         2
2       A  7       IV         3
3       A  8        V         4
4       A  8       VI         5
5       A 19      VII         6
6       A  8     VIII         7
7       A  1       IX         8
8       B  6        I         0
9       B  9       II         1
10      B 12      III         2
11      B 11       IV         3
12      B  7        V         4
13      B  6       VI         5
14      B  1      VII         6
15      B  0     VIII         7
16      B  0       IX         8
17      C  1        I         0
18      C  1       II         1
19      C  6      III         2
20      C  8       IV         3
21      C 23        V         4
22      C  7       VI         5
23      C  5      VII         6
24      C  1     VIII         7
25      C  0       IX         8
26      D  1       IV         3
27      D  3        V         4
28      D  7       VI         5
29      D 14      VII         6
30      D 16     VIII         7
31      D 11       IX         8
>

Kjetil Halvorsen

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

```