thanks to Petr Pikal, Andy Liaw, Gérald Jean, Sam Buttrey, Patrick
Connolly and especially Nick Ellis for their quick and helpful answers.
the problem stems from the fact that the factor variable is string, i.e.
""s are not NAs.
dear s-plus-users,
thanks to david m. smith, renaud lancelot, d. mckenzie, gérald
jean, and nick
ellis for their quick answers. they all suggested the same solution:
leidenbeg.omiss <- leidenbeg[!is.na(leidenbeg$the.factor.variable),
]
still, the dimensions of the two datasets are the same:
> dim(leidenbeg)
[1] 686 894
> dim(leidenbeg.omiss)
[1] 686 894
my factor variable is THER (summary see below), so i wonder why the
new
dataset doesn't contain 37 rows fewer than the old one. probably obious
to
everybody except for me.
thanks, this is a great community.
bernd
*** Summary Statistics for data in: leidenbeg ***
THER
: 37
bt:207
pa:108
pd:334
Bernd Puschner schrieb:
> dear s-plus users,
>
> i have a hudge data set
>
> > dim(leidenbeg)
> [1] 686 889
>
> on part of which i want to perform survival analyses. i want to create
a
> data set which omits all cases with missings on the factor variable
of
> the survivals. if i compute
>
> leidenbeg.omiss <- na.omit (leidenbeg)
>
> s-splus kicks out cases with missings on any of the variables. there's
> probably a very easy solution to that.
>
> thanks a lot
>
> bernd
>
> ------------------------------------------------------------------------
> ---------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu.
To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu
with
> the BODY of the message: unsubscribe s-news
---------------------------------------------------------------------
This message was distributed by s-news@lists.biostat.wustl.edu.
To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
the BODY of the message: unsubscribe s-news
