A square matrix in which the entry in cell (j, k) is some measure of the similarity (or distance) between the items to which row j and column k correspond. A simple example would be a standard mileage chart—the smaller the entry, the closer together are the two items. Proximity matrices form the data for multidimensional scaling. Asymmetric matrices can occur (for example, if the measurement is time taken, then the journey from top to bottom of a hill will be shorter than the journey from bottom to top).
Subjects: Probability and Statistics.