[R] problem in WRS2

[Bert Gunter]

As I said previously, you probably should contact the maintainer. But a
(wild??) guess might be that robust/resistant procedures -- which is what
you are using -- can downweight data values to 0 weight, effectively
removing them.This might lead to effectively empty cells in a cross
tabulation that would then yield your error, despite the crosstab on raw
data showing all cells with data.

Again, note my weasel words ("might","could", etc.), so the above could be
complete nonsense. Which is why I suggested you contact the maintainer.

-- Bert

On Thu, Nov 7, 2019 at 7:50 AM greg holly <mak.hholly using gmail.com> wrote:

> Hi David, Jim and Bert;
>
> Thanks so much. Your responses are much appreciated. Here is the results
> when I create a cross-tabulation from xtab(Grup~Time,cp) for the whole
> data. It seems to me there is no problem. I am wondering why still I
> have the
>  Error in x[[grp[i]]] :
>   attempt to select less than one element in get1index
>
> ERROR when I am running the t2way(y ~ Grup*Time, data = cp)  using WRS2
> package. Yours advice are highly appreciated.
>
> Kind regards,
> Greg
>
>
>         Time
> Grup     Cp0      Cp1      Cp14     Cp21     Cp28     Cp3      Cp7
> 1             25      25      25      25      25      25      25     175
>            14.29   14.29   14.29   14.29   14.29   14.29   14.29       -
>             62.5    62.5    62.5    62.5    62.5    62.5    62.5    62.5
>
> 2             15      15      15      15      15      15      15     105
>            14.29   14.29   14.29   14.29   14.29   14.29   14.29       -
>             37.5    37.5    37.5    37.5    37.5    37.5    37.5    37.5
>
>               40      40      40      40      40      40      40     280
>            14.29   14.29   14.29   14.29   14.29   14.29   14.29  100.00
>
>
>
>
>
> On Wed, Nov 6, 2019 at 7:46 PM David Winsemius <dwinsemius using comcast.net>
> wrote:
>
> > If a crosstabs on the two factors has any zero counts it might explain.
> >
> > —
> > David
> >
> > Sent from my iPhone
> >
> > > On Nov 6, 2019, at 5:42 PM, Jim Lemon <drjimlemon using gmail.com> wrote:
> > >
> > > Hi Greg,
> > > I tried this:
> > >
> > > cp<-read.table(text="Birey Grup Time y
> > > 1    1  Cp1 0.7916386
> > > 1    1  Cp3 1.7463777
> > > 1    1  Cp7 1.2008390
> > > 1    1 Cp14 0.6311380
> > > 1    1 Cp21 2.1563557
> > > 1    1 Cp28 1.2008390",
> > > header=TRUE)
> > > library(prettyR)
> > > xtab(Grup~Time,cp)
> > > Crosstabulation of Grup by Time
> > >       Time
> > > Grup     Cp1      Cp14     Cp21     Cp28     Cp3      Cp7
> > > 1              1       1       1       1       1       1       6
> > >          16.67   16.67   16.67   16.67   16.67   16.67       -
> > >            100     100     100     100     100     100     100
> > >
> > >              1       1       1       1       1       1       6
> > >          16.67   16.67   16.67   16.67   16.67   16.67  100.00
> > >
> > > Maybe doing the same on your entire data set will reveal something.
> > >
> > > Jim
> > >
> > >> On Thu, Nov 7, 2019 at 8:58 AM greg holly <mak.hholly using gmail.com>
> wrote:
> > >>
> > >> I got the following error message after running  t2way(y ~ Grup*Time,
> > data
> > >> = cp)
> > >>  Error in x[[grp[i]]] :
> > >>  attempt to select less than one element in get1index
> > >>
> > >> a part of the data is given below. Your help is highly appreciated.
> > >>
> > >> Greg
> > >>
> > >>> head(cp)
> > >>  Birey Grup Time         y
> > >> 1     1    1  Cp1 0.7916386
> > >> 2     1    1  Cp3 1.7463777
> > >> 3     1    1  Cp7 1.2008390
> > >> 4     1    1 Cp14 0.6311380
> > >> 5     1    1 Cp21 2.1563557
> > >> 6     1    1 Cp28 1.2008390
> > >>
> > >>        [[alternative HTML version deleted]]
> > >>
> > >> ______________________________________________
> > >> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > >> 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.
> > >
> > > ______________________________________________
> > > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > > 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.
> >
> >
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
>

	[[alternative HTML version deleted]]