[R] Adding across columns ignoring NA

[Bert Gunter]

Berend:

No. The OP's "variables" appear to be vectors. sum() sums columnwise
(vectorwise).

Something like

rowSums(cbind(var1,var2,var3,var4,var6), na.rm=TRUE)

appears to be what is wanted. But of course,Jin Choi appears to need
to do some homework -- e.g. by reading the Introduction to R online
manual to gain a better understanding of R's standard data structures.

Cheers,
Bert



On Fri, Jul 19, 2013 at 10:06 PM, Berend Hasselman <bhh at xs4all.nl> wrote:
>
> On 20-07-2013, at 06:37, Jin Choi <oohpsjin at gmail.com> wrote:
>
>> I am having difficulty finding a solution to devising an R code to do the
>> following:
>>
>> I have 5 numerical variables and I would like to create a new variable that
>> is the sum of those 5 variables. However, there are many NA values
>> throughout these 5 variables and everytime I run the following code
>>
>> new_variable=var1+var2+var3+var4+var5
>>
>> I get NA as the sum whenever one of those 5 variables are NA. I cannot
>> figure out a way to have new_variable represent the sum for only those
>> values that are not NA.
>>
>> As an example,
>> if var1=3
>> var2=3
>> var3=NA
>> var4=NA
>> var5=2
>>
>> I would like new_variable to be 8 but I keep getting NA and I have
>> unsuccessfully tried different methods to do so. I feel there is a simple
>> method to solve my problem but I am unaware of such. I would appreciate any
>> guidance!
>
> ?sum
>
> Have a look at the na.rm argument of sum.
>
> Berend
>
> ______________________________________________
> 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.



-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm