# [R] robust method to obtain a correlation coeff?

Martin Maechler maechler at stat.math.ethz.ch
Tue Aug 25 12:11:44 CEST 2009

```>>>>> "JK" == John Kane <jrkrideau at yahoo.ca>
>>>>>     on Mon, 24 Aug 2009 10:01:56 -0700 (PDT) writes:

JK> I may be misunderstanding the question but would cor(d1,
JK> use='complete.obs') or some other variant of "use" help?

Yes!  I've been a bit appalled to read that nobody else gave
this (in my view  *the correct* !) answer.

Leaving away NA's via na.omit() or similar is more general, but
for cor() definitely less flexible!

Martin Maechler, ETH Zurich

JK> --- On Mon, 8/24/09, Christian Meesters
JK> <meesters at imbie.uni-bonn.de> wrote:

>> From: Christian Meesters <meesters at imbie.uni-bonn.de>
>> Subject: [R] robust method to obtain a correlation coeff?
>> To: "r-help at r-project.org Help" <r-help at r-project.org>
>> Received: Monday, August 24, 2009, 10:47 AM Hi,
>>
>> Being a R-newbie I am wondering how to calculate a
>> correlation coefficient (preferably with an associated
>> p-value) for data like:
>>
>> > d[,1] [1] 25.5 25.3 25.1   NA 23.3 21.5 23.8 23.2 24.2
>> 22.7 27.6 24.2 ...  > d[,2] [1]  0.0 11.1  0.0   NA  0.0
>> 10.1 10.6  9.5  0.0 57.9  0.0  0.0  ...
>>
>> Apparently corr(d) from the boot-library fails with NAs
>> in the data, also cor.test cannot cope with a different
>> number of NAs.  Is there a solution to this problem
>> (calculating a correlation coefficient and ignoring
>> different number of NAs), e.g. Pearson's corr coeff?
>>
>> If so, please point me to the relevant piece of
>> documentation.
>>
>> TIA Christian
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> http://www.R-project.org/posting-guide.html and provide
>> commented, minimal, self-contained, reproducible code.
>>

__________________________________________________________________
JK> Be smarter than spam. See how smart SpamGuard is ons in
JK> Mail and switch to New Mail today or register for free
JK> at http://mail.yahoo.ca

JK> ______________________________________________
JK> R-help at r-project.org mailing list