[R] R square for Monotone regression

vito muggeo vmuggeo at dssm.unipa.it
Wed Feb 20 09:17:58 CET 2008


Dear Thierry,
I do not know fdrtool::monoreg(), however I think that the simple R^2 is 
meaningless in thise case. An unconstrained (non-monotonic) fit will 
have always a higher R^2 and more degrees-of-freedom.

A better approach would use any goodness-of-fit criterion penalized for 
the degrees of freedom of the fit itself (for instance AIC, BIC, GCV,..)

Hope this helps you,

vito


thierrydb ha scritto:
> I'm using the monoreg function (with weights) from the fdrtool package.  
> How can I calculate the R square for this type of regression?
> 
> 
> Thanks for your help,
> 
> 
> Thierry

-- 
====================================
Vito M.R. Muggeo
Dip.to Sc Statist e Matem `Vianelli'
Università di Palermo
viale delle Scienze, edificio 13
90128 Palermo - ITALY
tel: 091 6626240
fax: 091 485726/485612



More information about the R-help mailing list