This is known as the calibration problem: given data from a regression
of y on x and an observation y*, make inferences about the
corresponding x*.
Osborne, C. (1991). Statistical calibration: a review. Internat.
Statist. Rev. 59 309-336
may be a good article to start your reading on this with - no doubt
others will mention their favourite articles.
Murray Jorgensen
Zoann Nugent wrote:
> I am a confused biometrician.
>
> A student has brought me some data: a set of dilutions of a chemical and the absorbance for each dilution. She wants to make a standard regression with
> Y as the fixed dilutions and X as the absorbances, then use the line to calculate concentration from absorbance.
>
> Can I do a Model I linear regression on the (log transformed) data and make a line from Ys measured without error and X measured with error, then use
> it to assign errors to the Y values I estimate?
>
> This is Chemistry 101 but I cannot find an explanation or guide for the stats.
--
Dr Murray Jorgensen http://www.stats.waikato.ac.nz/Staff/maj.html
Department of Statistics, University of Waikato, Hamilton, New Zealand
Email: [log in to unmask] [log in to unmask] Fax 7 838 4155
Phone +64 7 838 4773 wk Home +64 7 825 0441 Mobile 021 0200 8350
|