[R] Using eval in multinom argument

[Charles C. Berry]

Forget eval(parse(text = ))

See

 	?as.formula
 	?update.formula

and try out the example() s there.

HTH,

Chuck




On Mon, 16 Feb 2009, Crouch, Daniel wrote:

> Hi,
>
> I am having difficulty entering a 'programmable' argument into the multinom function from the nnet package. Interactively, I can get the function to work fine by calling it this way:
>
> z1=multinom(formula = class.ind(grp[-outgroup])~ (PC1 + PC2 + PC3), data=data.frame(scores))
>
> However I need to be able to change the number of variables I am looking for in 'scores' and so am trying to call it this way...
>
> z1=multinom(formula = class.ind(grp[-outgroup])~ eval(parse(text=PCnames)), data=data.frame(scores))
>
> ...where, for example, PCnames = c("PC1", "+",   "PC2", "+",   "PC3")
>
> This gives no error messages, but only the last variable (in this case PC3) gets considered in the model. z1 looks like this:
>
>  (Intercept) eval(parse(text = PCnames))
> 2    3.530352                  -116.87140
> 3   -1.308613                    13.59134
> 4    3.662172                   -57.52198
> 5   -1.216041                  -242.38827
> 6   -9.377894                  -367.71614
> 7   -3.145738                  -286.19766
>
> Rather than this:
>
>  (Intercept)       PC1        PC2       PC3
> 2   288.97131   889.281  3776.5837 -2105.751
> 3  -712.53519  2775.663  8490.5724  8602.834
> 4   229.17772  4234.950   329.6995 -2182.238
> 5    85.54585 -3036.657  3968.2517 -3450.070
> 6  -676.55377 -9545.785  2422.5340 -7183.686
> 7  -631.91921 10997.432 -3310.2905 -5348.513
>
> Any help would be much appreciated.
>
> Thanks,
> Daniel Crouch
>
>
> Daniel Crouch
> Research Student
> Department of Medical & Molecular Genetics
> King's College London
> 8th Floor, Tower Wing
> Guy's Hospital
> London SE1 9RT
> United Kingdom
> ______________________________________________
> 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.
>

Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901