[R] box.Cox.powers() warning

[John Fox]

Dear Clara,

The warning is nothing to worry about. If start values aren't specified,
box.cox.powers() uses optimize() to find start values for each
transformation parameter (in this case, there's only one) prior to using
optim() to maximize the (possibly multivariate) likelihood. In the course of
finding start values, optimize() apparently tried an unreasonable value. In
this case, box.cox.powers(income, start=1) gives you the same result minus
the warning.

Regards,
 John

--------------------------------
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of cmhcordei at gmail.com
> Sent: February-06-08 1:05 PM
> To: R-help at r-project.org
> Subject: [R] box.Cox.powers() warning
> 
> Dear Rlist,
> 
> Using an example in box.cox.powers() help, I have the following warning
> message.
> 
> example:
> library(car)
> >attach(Prestige)
> 
> > box.cox.powers(income)
> Box-Cox Transformation to Normality
> 
>  Est.Power Std.Err. Wald(Power=0) Wald(Power=1)
>     0.1793   0.1108        1.6179       -7.4062
> 
> L.R. test, power = 0:  2.7103   df = 1   p = 0.0997
> L.R. test, power = 1:  47.261   df = 1   p = 0
> Warning message:
> In optimize(f = function(lambda) univ.neg.kernel.logL(x = X[, j],  :
>   NA/Inf replaced by maximum positive value
> 
> I do not know the reason for such warning message.
> Can someone help me on this matter?
> 
> Best regards,
> Clara Cordeiro
> 
> > sessionInfo()
> R version 2.6.1 (2007-11-26)
> i386-pc-mingw32
> 
> locale:
> LC_COLLATE=Portuguese_Portugal.1252;LC_CTYPE=Portuguese_Portugal.1252;L
> C_MONETARY=Portuguese_Portugal.1252;LC_NUMERIC=C;LC_TIME=Portuguese_Por
> tugal.1252
> 
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
> 
> other attached packages:
> [1] nnet_7.2-38    car_1.2-7      MASS_7.2-38    FitAR_1.0
> leaps_2.7
> [6] lattice_0.17-2
> 
> loaded via a namespace (and not attached):
> [1] grid_2.6.1  tools_2.6.1
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.