R MixSim负重叠概率为负

时间:2020-02-03 22:10:15

标签: r overlap gaussian normal-distribution

我正试图估计观察值将被分类为两个可能的多元法线之一的可能性。使用R的MixSim软件包,这是我的设置和输出。

> Mu
          [,1]     [,2]     [,3]   [,4]
[1,] 1.071975 17.81168 122934.80 186201.2
[2,] 1.127719 23.85656  97991.84 211144.2
> S
, , 1

              [,1]          [,2]           [,3]          [,4]
[1,]    0.01025366      0.134281 -4.292023e+02  4.292023e+02
[2,]    0.13428099      5.066837 -1.220579e+04  1.220579e+04
[3,] -429.20227492 -12205.788493  3.213470e+07 -3.213470e+07
[4,]  429.20227492  12205.788493 -3.213470e+07  3.213470e+07

, , 2

              [,1]          [,2]          [,3]          [,4]
[1,]  6.638739e-03 -7.368134e-02  1.881672e+02 -1.881672e+02
[2,] -7.368134e-02  2.913884e+01 -7.487336e+04  7.487336e+04
[3,]  1.881672e+02 -7.487336e+04  2.039656e+08 -2.039656e+08
[4,] -1.881672e+02  7.487336e+04 -2.039656e+08  2.039656e+08

> 
> overlap(Pi=c(0.5, 0.5), Mu=Mu, S=S)
$OmegaMap
     [,1] [,2]
[1,]  1.0   -1
[2,]  0.5    1

$BarOmega
[1] -0.5

$MaxOmega
[1] 0

$rcMax
[1] 1 1

为什么OmegaMap的值为负?

0 个答案:

没有答案