Passing-Baklok regression would be one good nonparametric alternative
instead of linear regression when the independent variable is not free of
error.
The method is detailly described in:
Passing H, Bablok W: A new biometrical procedure for testing the Equality of
Measurements from two different analytical
methods.J.Clin.Chem.Clin.Biochem.,Vol21,1983, pp. 709-720.
Another excellent reference comparing different methods:
Passing H, Bablok W: Comparison of several regression procedures for method
comparison studies and determination of sample sizes.
J.Clin.Chem.Clin.Biochem., Vol.22,1984,pp.431-445.
A user friendly software (Excel-Add-in) which provides the methods is
Analyse-it Clinical Laboratory.
http://www.analyse-it.com/
I hope this helps.
Bernd
\\||//
( .. )
---------------------oOO--( )--OOo-----------------------
Dipl.-Ing. Bernd Genser
Clinical Research Statistician
Department of Medical and Chemical Laboratory Diagnosis
BL II, University Hospital Graz
Auenbruggerplatz 15
8036 Graz
Austria
Tel.: +43(316)385-2420
Fax: +43(316)385-3430
<mailto:[log in to unmask]>
----------------------------------------------------------
|