[R] cor

[Prof Brian Ripley]

What version of R?  From the NEWS file for 1.8.1 alpha:

    o   cor(<matrix>, use = "pairwise") gave wrong result in 1.8.0 (only).
        (PR#4646)

On Tue, 11 Nov 2003, Peter B. Mandeville wrote:

> Greetings:
> 
> It would seem to me that the three arguments "a", "c" and "p" ought to 
> produce the same result with a data frame that doesn't have any missing 
> data which is not the case. What am I doing wrong or what don't I understand?
> 
>  > tmp <- data.frame(vol,con,mot,esp,pri,sec)
>  > tmp
>      vol con mot esp pri sec
> 1  1.00 120  90  78  21   1
> 2  0.50  60  25  80  15   5
> 3  0.50  50  40  52  46   2
> 4  1.00 257  60  78  20   2
> 5  0.50 150  85  85  11   4
> 6  0.50 113   5  79  18   3
> 7  0.50 117  60  67  30   3
> 8  1.00 193  60  74  23   3
> 9  0.50 139  60  50  46   4
> 10 0.50 144  70  67  31   2
> 11 1.00 127  80  47  41  12
> 12 0.50 200  80  42  40  18
> 13 0.75 106  40  58  33   9
> 14 0.50 127  30  61  28  11
> 15 0.50 200  50  51  45   4
> 16 1.00 229  80  70  27   3
> 17 0.75 207  50  62  28  10
> 18 0.25  63  30  60  35   5
> 19 1.00  58  80  67  29   4
>  > cor(tmp)
>              vol         con         mot         esp         pri         sec
> vol  1.00000000  0.36263681  0.52616642  0.21163185 -0.21149234 -0.09813069
> con  0.36263681  1.00000000  0.34665834  0.02016283 -0.07877831  0.12922310
> mot  0.52616642  0.34665834  1.00000000 -0.04678632  0.04038370  0.03675419
> esp  0.21163185  0.02016283 -0.04678632  1.00000000 -0.94112303 -0.60128659
> pri -0.21149234 -0.07877831  0.04038370 -0.94112303  1.00000000  0.29576002
> sec -0.09813069  0.12922310  0.03675419 -0.60128659  0.29576002  1.00000000
>  > cor(tmp,use="a")
>              vol         con         mot         esp         pri         sec
> vol  1.00000000  0.36263681  0.52616642  0.21163185 -0.21149234 -0.09813069
> con  0.36263681  1.00000000  0.34665834  0.02016283 -0.07877831  0.12922310
> mot  0.52616642  0.34665834  1.00000000 -0.04678632  0.04038370  0.03675419
> esp  0.21163185  0.02016283 -0.04678632  1.00000000 -0.94112303 -0.60128659
> pri -0.21149234 -0.07877831  0.04038370 -0.94112303  1.00000000  0.29576002
> sec -0.09813069  0.12922310  0.03675419 -0.60128659  0.29576002  1.00000000
>  > cor(tmp,use="c")
>              vol         con         mot         esp         pri         sec
> vol  1.00000000  0.36263681  0.52616642  0.21163185 -0.21149234 -0.09813069
> con  0.36263681  1.00000000  0.34665834  0.02016283 -0.07877831  0.12922310
> mot  0.52616642  0.34665834  1.00000000 -0.04678632  0.04038370  0.03675419
> esp  0.21163185  0.02016283 -0.04678632  1.00000000 -0.94112303 -0.60128659
> pri -0.21149234 -0.07877831  0.04038370 -0.94112303  1.00000000  0.29576002
> sec -0.09813069  0.12922310  0.03675419 -0.60128659  0.29576002  1.00000000
>  > cor(tmp,use="p")
>            vol         con        mot         esp         pri         sec
> vol 1.0000000  0.35898321 0.33585184  0.26418611  0.31817707  0.34311708
> con 0.3589832  0.99240991 0.34399506  0.02035910 -0.07698392  0.12891702
> mot 0.3358518  0.34399506 0.98480785  0.04365569  0.08719418  0.05687228
> esp 0.2641861  0.02035910 0.04365569  0.99985490 -0.85247286 -0.55912698
> pri 0.3181771 -0.07698392 0.08719418 -0.85247286  0.98445398  0.30189124
> sec 0.3431171  0.12891702 0.05687228 -0.55912698  0.30189124  0.95034447
> 
> Thank you very much,
> 
> Peter B.
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 
> 

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595