Steve,

According to my calculations, the dose and the response below are correlated
r=.98. As I have said since 1991, such high levels of correlation do not
leave enough residual variance to calculate CR.  This data is thus an
invalid test of CR
data.
>
>    dose resp
>  1   10  5.2
>  2   10  5.3
>  3   10  6.9
>  4   10  5.8
>  5   10  6.6
>  6   15 16.4
>  7   15 17.9
>  8   15 18.3
>  9   15 19.5
> 10   15 16.9
> 11   15 17.5
> 12   15 19.1
> 13   15 18.8
> 14   15 17.1
> 15   20 35.2
> 16   20 30.4
> 17   20 37.2
> 18   20 29.5
> 19   20 34.6
> 20   20 31.9
> 21   25 41.8
> 22   25 45.2
> 23   25 43.2
> 24   25 46.5
> 25   25 41.7

The last data set contains curvilinear data. CR is not made for such data. I
am not saying transformations can not be used. I have just never done any
work with them on CR. I will have to do some simulations before knowing what
to make of it.  I am particularly concerned about how you calculate the
residuals with this nonlinear data.

Can you imagine what people would say if you tried to test other statistics
with such small data sets?  Please try to find some data that fits the
assumptions.

Bill





.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to