A good approximation can be obtained algebraically
if the model is linear, otherwise ...
The only treatment that comes immediately to mind
is by Alec Brownlee in his 1960 book, Statistical
theory and methodology in science and engineering
-- Wiley. He referred to it as using the
regression line in reverse. It is a good book and
well worth the trouble to seek out a copy.
Robert Burrows wrote:
>
> Standard curves are prepared in the laboratory by measuring the response
> of some system at several levels of a factor. A model is fit to the data
> which is useful for estimating the response as a function of the factor
> level. However, the standard curve is often used to infer the level of the
> factor given some observed response. How does one calculate the standard
> error of the estimated factor level in this case?
>
> TIA,
>
> --
> Robert B. Burrows, PhD
> New England Biometrics
> [EMAIL PROTECTED]
--
Bob Wheeler --- (Reply to: [EMAIL PROTECTED])
ECHIP, Inc.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
