[R] How to get around heteroscedasticity with non-linear leastsquares in R?

[Berton Gunter]

And an added US$.02 is that the raw response (optical density, counts (large
numbers) of radio decay, fluorescence units, etc.) in dose response curves
often varies over several orders of magnitude, so that, in conformance to
John Tukey's "First Aid" suggestions, a log transformation or something
similar is often a standard prescription for fitting dose/response curves
(with the usual handwringing about whether the error is additive or
multiplicative; there is typically  some of both, as David Rocke's papers of
a decade or more ago argue).

-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
 
"The business of the statistician is to catalyze the scientific learning
process."  - George E. P. Box
 
 

> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Christian Ritz
> Sent: Wednesday, February 22, 2006 1:22 AM
> To: quin.wills at googlemail.com
> Cc: r-help at stat.math.ethz.ch; p.dalgaard at biostat.ku.dk
> Subject: Re: [R] How to get around heteroscedasticity with 
> non-linear leastsquares in R?
> 
> Hi Quin,
> 
> the package 'drc' on CRAN deals with modelling dose-response curves.
> 
> Moreover it allows adjustment for heterogeneity by means of 
> 
> 
>   transformation (Box-Cox transformation)
> 
>   modelling the variance as a power of the mean.
> 
> 
> See the package documentation for more features.
> 
> 
> Christian
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
>