Issue regarding mahalanobis distance function

Hi, i am currently working on my thesis and i am trying the mahalanobis distance function to get the answer. The problem is i have a matrix of 242 lines and 2 colums and. In the first columns i have different mean values and in second column i have the standard deviation. What i want is at each line(in a for loop) i want to compare the line standard deviation and mean with the first line. Ultimately having a vector of 1 column and 242 lines. Can someone tell me how this can be done.

I tried using the 'mahal(X,Y)' but it does not work when dimension of both vectors are the same.

Answer 1

[Geoff Hayes]

Marc - please clarify what you mean by i want to compare the line standard deviation and mean with the first line. If X is your 242x2 matrix, then isn't the comparison just

mDist = mahal(X(1,:), X(k, :));

where k is the kth row that you are comparing to the first row in the for loop?

[Marc Saade]

yes i tried that but it did not work. it give this error:

Error using mahal (line 38)

The number of rows of X must exceed the number of columns.

Error in Untitled2 (line 37)

mDist(k) = mahal(s(1,:), s(k, :));

[Geoff Hayes]

Marc - from mahal it seems that the second input needs more rows than columns. Are you sure that you want to be using mahal?

Answer 2

[Marc Saade]

what is theire other that function, is theire a function that calculates the cos distance?

[Geoff Hayes]

are you just trying to find the nearest distance between the two sets of values? perhaps just use the squared distance? Something like

dist = sqrt(sum((X(1,:) -  X(k, :))^2));

Or would this not produce the results you need/expect?