R help - Dec 2002 - summary stats including NA's into new dataframe

List,

I am trying to extract summary statistics from a data frame with several
variables (and NAs) into a dataframe with the columns: Variablename (ie the
colnames of original data), mean, stdev, max, min, Valid N, Missing Values.

Extracting the statistics is straightforward using stack and aggregate.
However, I haven''t succeeded in obtaining the number of Missing Values.
I
can extract these from describe (Hmisc library), but surely there is a
simpler way similar to obtaining the mean using aggregate?

Suggestions are much appreciated
Thanks Herry

--------------------------------------------
Alexander Herr - Herry
CSIRO Sustainable Ecosystems:
http://www.cse.csiro.au/
--------------------------------------------


	[[alternate HTML version deleted]]

Alexander.Herr at csiro.au wrote:
> List,
> 
> I am trying to extract summary statistics from a data frame with several
> variables (and NAs) into a dataframe with the columns: Variablename (ie the
> colnames of original data), mean, stdev, max, min, Valid N, Missing Values.
> 
> Extracting the statistics is straightforward using stack and aggregate.
> However, I haven't succeeded in obtaining the number of Missing Values.
I
> can extract these from describe (Hmisc library), but surely there is a
> simpler way similar to obtaining the mean using aggregate?
The similar way is:

aggregate(......., function(x) sum(is.na(x)))

Uwe Ligges
> Suggestions are much appreciated

Thanks Uwe,
Can''t seem to get your formula to work...
I should have made this clearer. I am after a listing of the number of NAs
and Valid Ns (or total N)for export to csv,eg:
Variable, mean, Missing Values, Valid N
test,	6.00000,2,18
bummer,5.44444,1,19

from:

x<-c(1,4,2,6,8,3,5,6,7,8,7,2,4,7,5,1,8,9,8,9)
labl<-gl(2,2,length=20,labels=c("test","bummer"))
x[3]<-NA
x[5]<-NA
x[6]<-NA


aggregate(x,by=list(labl),mean, sum(is.na(x)))
#  Group.1  x
#1    test NA
#2  bummer NA

aggregate(x,by=list(labl),mean, na.rm=T)
#  Group.1           x
#1    test 6.000000
#2  bummer 5.444444

aggregate(x,by=list(labl),sum(is.na(x)))
# Error in FUN(X[[1]], ...) : Argument "INDEX" is missing, with no
default



Cheers Herry


--------------------------------------------
Alexander Herr - Herry
Northern Futures
Davies Laboratory
PMB, Aitkenvale, QLD 4814
Phone (07) 4753 8510
Fax   (07) 4753 8650
Home: http://batcall.csu.edu.au/~aherr
CSIRO Sustainable Ecosystems:
http://www.cse.csiro.au/
--------------------------------------------



-----Original Message-----
From: Uwe Ligges [mailto:ligges@statistik.uni-dortmund.de]
Sent: Wednesday, 18 December 2002 5:30 PM
To: Alexander.Herr@csiro.au
Cc: r-help@stat.math.ethz.ch
Subject: Re: [R] summary stats including NA''s into new dataframe


Alexander.Herr@csiro.au wrote:
> List,
> 
> I am trying to extract summary statistics from a data frame with several
> variables (and NAs) into a dataframe with the columns: Variablename (ie
the
> colnames of original data), mean, stdev, max, min, Valid N, Missing
Values.
> 
> Extracting the statistics is straightforward using stack and aggregate.
> However, I haven''t succeeded in obtaining the number of Missing
Values. I
> can extract these from describe (Hmisc library), but surely there is a
> simpler way similar to obtaining the mean using aggregate?
The similar way is:

aggregate(......., function(x) sum(is.na(x)))

Uwe Ligges
> Suggestions are much appreciated

	[[alternate HTML version deleted]]

I think my dstats function does what you want, if I understand you coorrectly.
 You could apply it over rows or columns:

  http://home.earthlink.net/~bmagill/MyMisc.html

It is there along with several other functions.



On Thu, 19 Dec 2002 11:18:25 +1000 Alexander.Herr at csiro.au wrote:
> Thanks Uwe,
> Can't seem to get your formula to work...
> I should have made this clearer. I am after a
> listing of the number of NAs
> and Valid Ns (or total N)for export to csv,eg:
> Variable, mean, Missing Values, Valid N
> test,	6.00000,2,18
> bummer,5.44444,1,19
> 
> from:
> 
> x List,
> > 
> > I am trying to extract summary statistics
> from a data frame with several
> > variables (and NAs) into a dataframe with the
> columns: Variablename (ie
> the
> > colnames of original data), mean, stdev, max,
> min, Valid N, Missing
> Values.
> > 
> > Extracting the statistics is straightforward
> using stack and aggregate.
> > However, I haven't succeeded in obtaining the
> number of Missing Values. I
> > can extract these from describe (Hmisc
> library), but surely there is a
> > simpler way similar to obtaining the mean
> using aggregate?
> 
> The similar way is:
> 
> aggregate(......., function(x) sum(is.na(x)))
> 
> Uwe Ligges
> 
> > Suggestions are much appreciated
> 
> 
> 	[[alternate HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-help
>