Question about subset

[sbihorel]

Hi,

The help page for subset states "subset: logical expression indicating elements or rows to keep: missing values are taken as false."

Before I try to re-invent the wheel, I would like to know if one of the base or recommended packages would contain a variant of the subset function that would consider missing values as true.

Thanks

______________________________________________
[hidden email] 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.

[Bert Gunter-2]

Please read the "Details" section of the help file -- it essentially
tells you what to do by directly using indexing; e.g. for a vector,
index by:
[is.na(subset) | subset]  .

Adjust as necessary for your data structure.Or look at the code of,
e.g. subset.data.frame and create your own subset version/method by
making the trivial modification to that code.

R is open source. Sometimes it's worthwhile to take advantage of this
by making a simple modification to its code on your own before hunting
for packages.

Cheers,
Bert



Cheers,
Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Sun, Apr 8, 2018 at 9:06 PM, Sebastien Bihorel
<[hidden email]> wrote:

> Hi,
>
> The help page for subset states "subset: logical expression indicating elements or rows to keep: missing values are taken as false."
>
> Before I try to re-invent the wheel, I would like to know if one of the base or recommended packages would contain a variant of the subset function that would consider missing values as true.
>
> Thanks
>
> ______________________________________________
> [hidden email] 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.

______________________________________________
[hidden email] 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.

[David Winsemius]

Sent from my iPhone

> On Apr 8, 2018, at 9:06 PM, Sebastien Bihorel <[hidden email]> wrote:
>
> Hi,
>
> The help page for subset states "subset: logical expression indicating elements or rows to keep: missing values are taken as false."
>
> Before I try to re-invent the wheel, I would like to know if one of the base or recommended packages would contain a variant of the subset function that would consider missing values as true.
>
Just use a Boolean expression

is.na(col)|(col==0)

> Thanks
>
> ______________________________________________
> [hidden email] 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.

______________________________________________
[hidden email] 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.

[sbihorel]

Thanks.

That works great!

> df <- data.frame(x=c(1,1,NA,NA,2), y=c('a','a','a','b','b'), z=c(TRUE,FALSE,TRUE,FALSE,TRUE))
> cond1 <- 'x==1'
> cond2 <- 'x==1 & z'
> df
   x y     z
1  1 a  TRUE
2  1 a FALSE
3 NA a  TRUE
4 NA b FALSE
5  2 b  TRUE
> subset(df, subset = ifelse(is.na(eval(parse(text=cond1))), TRUE, eval(parse(text=cond1))))
   x y     z
1  1 a  TRUE
2  1 a FALSE
3 NA a  TRUE
4 NA b FALSE
> subset(df, subset = ifelse(is.na(eval(parse(text=cond2))), TRUE, eval(parse(text=cond2))))
   x y    z
1  1 a TRUE
3 NA a TRUE


----- Original Message -----
From: "S Ellison" <[hidden email]>
To: "Sebastien Bihorel" <[hidden email]>
Sent: Monday, April 9, 2018 8:31:55 AM
Subject: RE: Question about subset

> Before I try to re-invent the wheel, I would like to know if one of the base or
> recommended packages would contain a variant of the subset function that
> would consider missing values as true.

The subset argument to subset is something that evaluates to logical - a vector of True or False.
If you wanted to treat missing values _in that_ as TRUE, wrap it in an ifelse:

ifelse(is.na(condition), TRUE, condition)

where condition can be a logical or an expression evaluating to logical.

S Ellison



*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}

______________________________________________
[hidden email] 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.

[sbihorel]

Thanks.

S. Elison provided a similar but apparently more general solution (see other post in thread).


----- Original Message -----
From: "David Winsemius" <[hidden email]>
To: "Sebastien Bihorel" <[hidden email]>
Cc: [hidden email]
Sent: Monday, April 9, 2018 12:33:41 AM
Subject: Re: [R] Question about subset

Sent from my iPhone

> On Apr 8, 2018, at 9:06 PM, Sebastien Bihorel <[hidden email]> wrote:
>
> Hi,
>
> The help page for subset states "subset: logical expression indicating elements or rows to keep: missing values are taken as false."
>
> Before I try to re-invent the wheel, I would like to know if one of the base or recommended packages would contain a variant of the subset function that would consider missing values as true.
>
Just use a Boolean expression

is.na(col)|(col==0)

> Thanks
>
> ______________________________________________
> [hidden email] 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.

______________________________________________
[hidden email] 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.