[R] analysis of residual sum of squares in R

Sebastian Luque spluque at gmail.com
Wed Feb 15 00:59:07 CET 2006


Dear R-helpers,

There's a method named as in this message's subject that has been proposed
by:

Chen Y, Jackson DA, Harvey HH (1992) A comparison of von Bertalanffy and
polynomial functions in modelling fish growth data.  Canadian Journal of
Fisheries and Aquatic Sciences 49:1228-1235.

which is used to compare a group of non linear models.  The method tests
whether all the fitted curves describe the same population of data.  The
test is an F ratio:

F = (RSSp - sum(RSSi) / DFp - sum(DFi)) / (sum(RSSi) / sum(DFi))

where the RSS and DF are the residual sum of squares and degrees of
freedom, respectively.  The 'p' and 'i' subscripts denote the pooled or
each individual curve, respectively.  If this procedure has already been
implemented in a package, I'd be happy to hear about it.  A search of the
archives and other sources didn't help.

Cheers,

-- 
Sebastian Luque




More information about the R-help mailing list