Re: [Gretl-users] about mahalanobis distance calculation in gretl
I think in R it's the squared distance, and casually looking at your
results seems to confirm this.
Actually there seems to be a typo in the manual, Mahalonobis instead of
Mahalanobis.
-sven
Baris Altayligil schrieb:
Hello,
I want to ask a question about the mahalonobis distances calculation in
gretl. I calculated the distances in Mathematica 7.0 and R, they are the
same but totally different from gretl calculations. I am giving the data
and the results in the below. Is this a bug or something i do not know
about gretl calculation ?
DATA
v1 v2
1 13.0 20
2 14.0 25
3 12.5 40
4 12.0 22
5 12.5 33
6 12.0 35
7 11.0 21
8 10.0 25
9 10.0 42
10 12.0 30
11 10.5 35
12 10.0 28
13 11.0 25
14 9.0 40
15 11.5 33
Mahalanobis distances calculated in R and Mathematica
2.5713469
3.7042152
3.6085838
1.3058809
1.1389734
0.9227402
2.2294228
2.3743239
2.8943684
0.2110974
0.6510297
1.5641416
0.8572523
3.7788747
0.1877488
Mahalanobis distances calculated in gretl
1.603542
1.924634
1.899627
1.142751
1.067227
0.960594
1.493125
1.540884
1.701284
0.459453
0.806864
1.250656
0.925879
1.943933
0.433300
Assistant Prof. Baris Altayligil
Department of Economics
Istanbul University / Turkey
------------------------------------------------------------------------
Windows Live Hotmail: Your friends can get your Facebook updates, right
from Hotmail®.
<http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/...
------------------------------------------------------------------------
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users