R help - Oct 2014 - {car} outlierTest looses p/q values

Hi guys,

I came across a strange phenomena and can't figure out why it happens by 
myself so here we go.

I got a dataframe which consists of double numbers which I want to 
check, row-wise if there are outliers in the rows.

So I iterate over the rows and create a glm using the numbers of that 
particular row. Which might look like this:

case1)
         x1        x2        x3        x4        x5        x6 x7        
x8        x9        x10        x11
     0.00     3.91     0.00     0.00     0.00   68.03   40.39 0.00     
0.00      0.00       4.11

or like this:
case2)
         x1        x2        x3        x4        x5        x6 x7        
x8        x9        x10        x11
      1.00     1.00    1.00     1.00     1.00     1.00     1.00 1.00     
1.00     1.00      5.34

or any other combination of double numbers...

however, using a glm like this:

glModel <- glm(vector ~ some_other_meta_data_which_is_double_numbers)

and testing it with:

  test.Res <- outlierTest(glModel,digits=4,cutoff=Inf,n.max=Inf)

I always get a result consisting of the desired p and q values but not 
if the vector I use looks like case2. There is no error message and the 
computation does not stop either.
However, all p and q values are produced except for the last value x11.

Any idea why this particular value gets dropped from the output of the 
outlierTest Method in the car package.

Here is the sessioninfo:

  sessionInfo()
R version 3.1.1 (2014-07-10)
Platform: x86_64-redhat-linux-gnu (64-bit)

locale:
  [1] LC_CTYPE=en_US.utf8       LC_NUMERIC=C
  [3] LC_TIME=en_US.utf8        LC_COLLATE=en_US.utf8
  [5] LC_MONETARY=en_US.utf8    LC_MESSAGES=en_US.utf8
  [7] LC_PAPER=en_US.utf8       LC_NAME=C
  [9] LC_ADDRESS=C              LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods base

other attached packages:
[1] ggplot2_1.0.0      car_2.0-21         RColorBrewer_1.0-5 iNEXT_1.0
[5] vegan_2.0-10       lattice_0.20-29    permute_0.8-3

loaded via a namespace (and not attached):
  [1] colorspace_1.2-4 compiler_3.1.1   digest_0.6.4 grid_3.1.1
  [5] gtable_0.1.2     labeling_0.3     MASS_7.3-33 munsell_0.4.2
  [9] nnet_7.3-8       plyr_1.8.1       proto_0.3-10 Rcpp_0.11.2
[13] reshape2_1.4     scales_0.2.4     stringr_0.6.2    tools_3.1.1

Any help is highly appreciated.

Thanks

Phil

Dear Phil,

After reading your posting several times, I still don't understand what you
did. As usual, having a reproducible example illustrating the error would be a
great help. I do have a guess about the source of the error: glm() failed in
some way for the problematic case.

Best,
 John

------------------------------------------------
John Fox, Professor
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/
	

On Fri, 17 Oct 2014 12:53:30 +0200
 Phil <chobophil at gmail.com> wrote:
> Hi guys,
> 
> I came across a strange phenomena and can't figure out why it happens
by myself so here we go.
> 
> I got a dataframe which consists of double numbers which I want to check,
row-wise if there are outliers in the rows.
> 
> So I iterate over the rows and create a glm using the numbers of that
particular row. Which might look like this:
> 
> case1)
>          x1        x2        x3        x4        x5        x6 x7        
> x8        x9        x10        x11
>      0.00     3.91     0.00     0.00     0.00   68.03   40.39 0.00     
> 0.00      0.00       4.11
> 
> or like this:
> case2)
>          x1        x2        x3        x4        x5        x6 x7        
> x8        x9        x10        x11
>       1.00     1.00    1.00     1.00     1.00     1.00     1.00 1.00     
> 1.00     1.00      5.34
> 
> or any other combination of double numbers...
> 
> however, using a glm like this:
> 
> glModel <- glm(vector ~ some_other_meta_data_which_is_double_numbers)
> 
> and testing it with:
> 
>   test.Res <- outlierTest(glModel,digits=4,cutoff=Inf,n.max=Inf)
> 
> I always get a result consisting of the desired p and q values but not if
the vector I use looks like case2. There is no error message and the computation
does not stop either.
> However, all p and q values are produced except for the last value x11.
> 
> Any idea why this particular value gets dropped from the output of the
outlierTest Method in the car package.
> 
> Here is the sessioninfo:
> 
>   sessionInfo()
> R version 3.1.1 (2014-07-10)
> Platform: x86_64-redhat-linux-gnu (64-bit)
> 
> locale:
>   [1] LC_CTYPE=en_US.utf8       LC_NUMERIC=C
>   [3] LC_TIME=en_US.utf8        LC_COLLATE=en_US.utf8
>   [5] LC_MONETARY=en_US.utf8    LC_MESSAGES=en_US.utf8
>   [7] LC_PAPER=en_US.utf8       LC_NAME=C
>   [9] LC_ADDRESS=C              LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C
> 
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods base
> 
> other attached packages:
> [1] ggplot2_1.0.0      car_2.0-21         RColorBrewer_1.0-5 iNEXT_1.0
> [5] vegan_2.0-10       lattice_0.20-29    permute_0.8-3
> 
> loaded via a namespace (and not attached):
>   [1] colorspace_1.2-4 compiler_3.1.1   digest_0.6.4 grid_3.1.1
>   [5] gtable_0.1.2     labeling_0.3     MASS_7.3-33 munsell_0.4.2
>   [9] nnet_7.3-8       plyr_1.8.1       proto_0.3-10 Rcpp_0.11.2
> [13] reshape2_1.4     scales_0.2.4     stringr_0.6.2    tools_3.1.1
> 
> Any help is highly appreciated.
> 
> Thanks
> 
> Phil
> 
> ______________________________________________
> 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.