I have two matrices which elements are distances, created from an anisotropic cost analysis algorithm, between sites/populations.

           A      B      C
         _                _
    A   |  0     DBA    DCA |
        |                  |
    B   | DAB     0     DCB |
        |                  |
    C   | DAC     DBC     0 |

However, the matrices are not symmetric in that the distance from A to B is different from the distance from B to A, or according to the above matrix DBA $\ne$ DAB

I know I could use Mantel's test, although it has been heavily criticised and its performance has never been assessed on asymmetric matrices. I thought about using Generalised Dissimilarity Modelling but I have only found examples with symmetric examples.

I would like to know:

1) Is Generalised Dissimilarity Modelling appropriate in my case?

and in case it is not,

2) Are other methods available to relate two asymmetric distance matrices?

  • 1
    $\begingroup$ What is your question? $\endgroup$ – whuber Apr 27 '17 at 13:11
  • $\begingroup$ I tried to improve the post, I apologise for the poor quality it had. $\endgroup$ – Julian Wittische Apr 27 '17 at 19:00
  • $\begingroup$ @whuber Any ideas? $\endgroup$ – Julian Wittische Nov 23 '19 at 21:34
  • $\begingroup$ I don't know what your question is. Could you describe how, if at all, this distance matrix was obtained from data and state what you are trying to discover from those data? $\endgroup$ – whuber Nov 23 '19 at 22:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.