[R] Stepwise Discriminant Analysis - greedy.wilks

[Uwe Ligges]

Please so not double post.

After personal communication it turned out that

cor(data_indiciN2$I022N2, data_indiciN2$I025N2)

was exactly 1. Hence you cannot use both of them in your model and have 
to skip one of them.

Therefore,

gw_obj <- greedy.wilks(gruppo ~ . - I025N2, data = data_indiciN2, niveau 
= 0.1)

works for me.

Best,
Uwe Ligges





On 20.06.2012 01:15, martanair wrote:
> I don't understand which is the problem
> Could you help me please?
> Thanks in advance
> Marta
>
>
>
>> str(data_indiciN2)
> 'data.frame':   200 obs. of  36 variables:
>   $ gruppo: Factor w/ 2 levels "0","1": 2 2 2 2 2 2 2 2 2 2 ...
>   $ I001N2: num  19.32 8.22 28.35 7.24 14.7 ...
>   $ I002N2: num  2.92 2.54 0.11 1.6 7.12 ...
>   $ I003N2: num  -22.362 -0.222 -19.291 -21.553 2.935 ...
>   $ I004N2: num  100 100 100 100 9.48 ...
>   $ I005N2: num  100 100 100 100 5.99 ...
>   $ I006N2: num  3.688 2.975 17.717 3.694 0.278 ...
>   $ I007N2: num  200 200 200 200 12.6 ...
>   $ I008N2: num  7.01 8.74 3.59 5.87 3.01 ...
>   $ I009N2: num  -18.75 -2.14 -1.29 -18.79 1.27 ...
>   $ I010N2: num  -25.75 -10.88 -4.88 -24.66 -1.74 ...
>   $ I011N2: num  3.12 3.68 1.26 2.34 1.96 ...
>   $ I012N2: num  2.85 3.46 1.26 2.24 1.33 ...
>   $ I013N2: num  2.85 3.46 1.26 2.24 1.33 ...
>   $ I014N2: num  0 0.796 0.576 0.248 1.187 ...
>   $ I015N2: num  -187.17 -37.73 1.61 -101.2 1.32 ...
>   $ I016N2: num  0.929 0.884 0.131 0.655 1.06 ...
>   $ I017N2: num  0.431 0.219 0.131 0.561 0.996 ...
>   $ I018N2: num  -0.9855 0.22449 0.43248 -0.00575 1.18704 ...
>   $ I019N2: num  -39.291 -1.346 -15.589 -22.193 0.485 ...
>   $ I020N2: num  -15.266 -5.536 0.503 -22.513 0.159 ...
>   $ I021N2: num  -31.9962 -3.4831 6.0837 -26.5905 0.0608 ...
>   $ I022N2: num  0.4771 1.5894 0.0827 0.8466 2.6159 ...
>   $ I023N2: num  0.37 0.219 0.409 0.562 0.684 ...
>   $ I024N2: num  100 100 -68.3 100 43.9 ...
>   $ I025N2: num  0.4771 1.5894 0.0827 0.8466 2.6159 ...
>   $ I026N2: num  100 79.94 100 100 2.84 ...
>   $ I027N2: num  0 0.4 0 0 2.2 ...
>   $ I028N2: num  -282.14 -50.56 -52.44 -7.79 28.54 ...
>   $ I029N2: num  -20.47 -1.74 7.87 -19.68 1.77 ...
>   $ I030N2: num  7.19 27.25 25.28 100 2.68 ...
>   $ I031N2: num  -100 -100 8.989 -100 0.598 ...
>   $ I032N2: num  75.4 67 59 0 95.2 ...
>   $ I033N2: num  17.45 5.74 6.74 100 1.49 ...
>   $ I034N2: num  0.472 0.815 0.482 0.399 1.557 ...
>   $ I035N2: num  0 1 100 100 100 100 100 100 0 100 ...
>> gw_obj<- greedy.wilks(gruppo~., data = data_indiciN2,niveau = 0.1)
> Errore in summary.manova(e2, test = "Wilks") : residuals have rank 2<  3
>
> --
> View this message in context: http://r.789695.n4.nabble.com/Stepwise-Discriminant-Analysis-greedy-wilks-tp4633902.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.