# mahalCorAna

mahalCorAna computes Mahalanobis distances (in squared units) for each row of matrix Y

## Syntax

• d=mahalCorAna(Y,MU)example

## Description

 d =mahalCorAna(Y, MU) Example of computation of MD.

## Examples

expand all

### Example of computation of MD.

Generate a contingency table.

nrow=30;
ncol=5;
% Fix the marginals of the two rows
nrowt=200*ones(1,nrow);
% Fix the marginals of the three columns
ncolt=1200*ones(ncol,1);
% Generate the contingency table
out=rcontFS(nrow,ncol,nrowt,ncolt);
N=out.m144;
MU=sum(N,1)/sum(nrowt);
% Compute MD ;
n=sum(N,'all');
P=N/n;
ProfileRows=P./sum(P,1);
d=mahalCorAna(ProfileRows,MU);

## Related Examples

expand all

### Find total inertial of contingency table Generate a contingency table.

nrow=30;
ncol=5;
% Fix the marginals of the two rows
nrowt=200*ones(1,nrow);
% Fix the marginals of the three columns
ncolt=1200*ones(ncol,1);
% Generate the contingency table
out=rcontFS(nrow,ncol,nrowt,ncolt);
N=out.m144;
n=sum(N,'all');
P=N/n;
r=sum(P,2); % row masses
c=sum(P,1); % centroid of row masses
ProfileRows=P./r;
d2=r.*mahalCorAna(ProfileRows,c);
% d2 is the total inertia
disp('Total inertia')
disp(sum(d2));

## Input Arguments

### Y — Input data. Matrix.

I x J Profile rows matrix; n observations and v variables.

Rows of Y represent observations, and columns represent variables.

Data Types: single|double

### MU — Centroid. Vector.

1 x J vector containing centroid and covariance matrix to use

Data Types: single| double

## Output Arguments

### d —Mahalanobis distances.  Vector

n x 1 vector which contains the squared Mahalanobid distances.

$d(i) = (y_i-\mu)^T \times diag(\mu)^{-1} \times (y_i-\mu), \qquad i=1, 2, \ldots, n$