[R] Remove missings (quick question)

Bert Gunter gunter.berton at gene.com
Fri Nov 9 18:23:01 CET 2012


Marc et. al:

On Fri, Nov 9, 2012 at 9:05 AM, Marc Schwartz <marc_schwartz at me.com> wrote:
> On Nov 9, 2012, at 10:50 AM, Eiko Fried <torvon at gmail.com> wrote:
>
>> A colleague wrote the following syntax for me:
>>
>> D = read.csv("x.csv")
>>
>> ## Convert -999 to NA
>> for (k in 1:dim(D)[2]) {
>>    I = which(D[,k]==-999)
>>    if (length(I) > 0) {
>>        D[I,k] = NA
>>    }
>> }
>>
>> The dataset has many missing values. I am running several regressions on
>> this dataset, and want to ensure every regression has the same subjects.
>>
>> Thus I want to drop subjects listwise for dependent variables y1-y9 and
>> covariates x1-x5 (if data is missing on ANY of these variables, drop
>> subject).
>>
>> How would I do this after running the syntax above?
>>
>> Thank you
>
>
> Modify the initial read.csv() call to:
>
>   D <- read.csv("x.csv", na.strings = "-999")
>
> That will convert all -999 values to NA's upon import so that you don't have to post-process it.
>
> See ?read.csv for more info.
>
> Once that is done, R's default behavior is to remove observations with any missing data (eg. NA values)
when using modeling functions.

This appears to be false. From ?lme (nlme package, nlme_3.1-105, R 2.15.2):

"na.action 	

a function that indicates what should happen when the data contain
NAs. The default action (na.fail) causes lme to print an error message
and terminate if there are any incomplete observations."

Frankly, I doubt that there is any uniformity for practically any
modeling options across the vast array of "modeling functions" in R
and (even recommended?) packages.

Cheers,
Bert

 Or you can pre-process using:
>
>   D.New <- na.omit(D)
>
> and then use D.New for all of your subsequent analyses. See ?na.omit.
>
> Regards,
>
> Marc Schwartz
>
> ______________________________________________
> 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




More information about the R-help mailing list