On Feb 4, 2011, at 7:06 PM, Gong-Yi Liao wrote:
Dear list:
I have tried MASS's mca function and SAS's PROC corresp on the
farms data (included in MASS, also used as mca's example), the
results are different:
R: mca(farms)$rs:
1 2
1 0.059296637 0.0455871427
2 0.043077902 -0.0354728795
3 0.059834286 0.0730485572
4 0.059834286 0.0730485572
5 0.012900181 -0.0503121890
6 0.038846577 -0.0340961617
7 0.005886752 -0.0438516465
8 -0.015108789 -0.0247221783
9 0.007505626 -0.0646608108
10 0.006631230 -0.0362117073
11 0.013309217 -0.0680733730
12 0.056549933 0.0010773359
13 0.015681958 0.0334320046
14 -0.065598990 0.0151619769
15 -0.046868229 0.0357782553
16 -0.003048803 0.0128157261
17 -0.051281437 0.0278941743
18 -0.051819085 0.0004327598
19 -0.072814626 0.0195622280
20 -0.072814626 0.0195622280
And in SAS's corresp output:
Row Coordinates
Dim1 Dim2
1 1.0607 -0.8155
2 0.7706 0.6346
3 1.0703 -1.3067
4 1.0703 -1.3067
5 0.2308 0.9000
6 0.6949 0.6099
7 0.1053 0.7844
8 -0.2703 0.4422
9 0.1343 1.1567
10 0.1186 0.6478
11 0.2381 1.2177
12 1.0116 -0.0193
13 0.2805 -0.5980
14 -1.1735 -0.2712
15 -0.8384 -0.6400
16 -0.0545 -0.2293
17 -0.9174 -0.4990
18 -0.9270 -0.0077
19 -1.3025 -0.3499
20 -1.3025 -0.3499
Is MASS's mca developed with different definition to SAS's
corresp ?
No, it's just that the values can only be defined up to a scaling
factor (the same situation as with eigenvector decompostion). Take a
look at the two dimensions, when each is put on the same scale:
> cbind(scale(rmca$D1),scale(smca$Dim1) )
[,1] [,2]
[1,] 1.2824421 1.28242560
[2,] 0.9316703 0.93168561
[3,] 1.2940701 1.29403231
[4,] 1.2940701 1.29403231
[5,] 0.2789996 0.27905048
[6,] 0.8401570 0.84016193
[7,] 0.1273161 0.12731705
[8,] -0.3267664 -0.32679513
[9,] 0.1623284 0.16237896
[10,] 0.1434174 0.14339716
[11,] 0.2878460 0.28787641
[12,] 1.2230376 1.22306216
[13,] 0.3391626 0.33913934
[14,] -1.4187467 -1.41879225
[15,] -1.0136458 -1.01364584
[16,] -0.0659382 -0.06588616
[17,] -1.1090928 -1.10915932
[18,] -1.1207208 -1.12076602
[19,] -1.5748033 -1.57475730
[20,] -1.5748033 -1.57475730
> cbind(scale(rmca$D2),scale(smca$Dim2) )
[,1] [,2]
[1,] 1.06673426 -1.06677626
[2,] -0.83006158 0.83012474
[3,] 1.70932841 -1.70932351
[4,] 1.70932841 -1.70932351
[5,] -1.17729983 1.17729909
[6,] -0.79784653 0.79781424
[7,] -1.02612383 1.02608072
[8,] -0.57849632 0.57844296
[9,] -1.51305605 1.51309282
[10,] -0.84735007 0.84739189
[11,] -1.59290964 1.59288798
[12,] 0.02520954 -0.02525321
[13,] 0.78230533 -0.78226073
[14,] 0.35478864 -0.35476797
[15,] 0.83720734 -0.83720166
[16,] 0.29988662 -0.29995785
[17,] 0.65272069 -0.65275711
[18,] 0.01012653 -0.01007904
[19,] 0.45775404 -0.45771681
[20,] 0.45775404 -0.45771681
--
David.
Thank you for any comments!
--
Gong-Yi Liao
Department of Statistics
University of Connecticut
--
David Winsemius, MD
West Hartford, CT
______________________________________________
R-help@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.
