[R-sig-eco] Function predict

[Gavin Simpson]

On Wed, 2010-07-07 at 20:44 -0300, Luciano Selzer wrote:
> Hi Manuel, your problem is that the new variables you created are of numeric
> type, while the model was fitted using factor. So the fix is to convert the
> new variables to factors so the type matchs

It may need more than that; not tested but I have often run into
problems where I had slightly different levels for factors in the
prediction data than in the fitting.

G

> Luciano
> 
> 
> 2010/7/7 Manuel Spnola <mspinola10 at gmail.com>
> 
> > Dear list members,
> >
> > I am fitting a logistic regression with 5 explanatory factors (which I
> > converted to factors):
> >
> > mod6 = glm(condicion ~ iluminacion + animales + cielo.raso + piso +
> > paredes, family=binomial, data=reglog)
> >
> > I want to obtain the predicted probabilities and the se using the function
> > "predict" for some combination of the factors.
> >
> > iluminacion = 1
> > animales = 0
> > cielo.raso = 0
> > piso = 0
> > paredes = 0
> >
> > newdata1 = data.frame(iluminacion, animales, cielo.raso, piso, paredes)
> >
> > newdata1$cond = predict(mod6, newdata = newdata1, type="response")
> >
> > > newdata1$cond = predict(mod6, newdata = newdata1, type="response")
> > Aviso en model.frame.default(Terms, newdata, na.action = na.action, xlev =
> > object$xlevels) :
> >  variable 'iluminacion' is not a factor
> > Aviso en model.frame.default(Terms, newdata, na.action = na.action, xlev =
> > object$xlevels) :
> >  variable 'animales' is not a factor
> > Aviso en model.frame.default(Terms, newdata, na.action = na.action, xlev =
> > object$xlevels) :
> >  variable 'cielo.raso' is not a factor
> > Aviso en model.frame.default(Terms, newdata, na.action = na.action, xlev =
> > object$xlevels) :
> >  variable 'piso' is not a factor
> > Aviso en model.frame.default(Terms, newdata, na.action = na.action, xlev =
> > object$xlevels) :
> >  variable 'paredes' is not a factor
> > Error: variables 'iluminacion', 'animales', 'cielo.raso', 'piso', 'paredes'
> > were specified with different types from the fit
> >
> > I don't understand what is wrong.
> > Any help will be appreciated.
> >
> > Best,
> >
> > Manuel
> >
> > --
> > Manuel Spnola, Ph.D.
> > Instituto Internacional en Conservacin y Manejo de Vida Silvestre
> > Universidad Nacional
> > Apartado 1350-3000
> > Heredia
> > COSTA RICA
> > mspinola at una.ac.cr
> > mspinola10 at gmail.com
> > Telfono: (506) 2277-3598
> > Fax: (506) 2237-7036
> >
> > _______________________________________________
> > R-sig-ecology mailing list
> > R-sig-ecology at r-project.org
> > https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> >
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%