Multidimensional Scaling

Complete implementation of non-metric multidimensional scaling. STATISTICA expects as input a similarity (e.g., correlation) or dissimilarity matrix.

Element Name Description
General
Detail of computed results reported Specifies the detail of computed results reported. If Minimal detail is requested, STATISTICA will report the final solution (configuration), plots of the final solution, and the Shepard diagram; if All results is requested, STATISTICA will report various other statistics, including the input (distance) matrix.
Number of dimensions Specifies the number of dimensions for the multidimensional scaling analysis.
Epsilon Specifies parameter Epsilon as the smallest distance that is to be considered important or significant by the program. All distances that are less than Epsilon will be considered to be 0 (zero) by the program. The smaller this value, the more iterations will be performed and the more precise is the solution. Change this value only if the iterative fitting procedure cannot converge (i.e., produce a solution) even after very many iterations.
Min. number of iterations Specifies the minimum number of iterations that are to be performed during the iterative fitting procedure. The minimum number of iterations is 6.
Max. number of iterations. Specifies the maximum number of iterations to be performed during the iterative fitting process.
Distance measure For raw data (non-matrix) input, specify the distance measure to compute; Multidimensional Scaling expects as input a matrix of similarities or dissimilarities (distance). When the input data do not specify a matrix file, i.e., are raw data, then you can select one of these distance measures to use as input into the analyses.