intour

This file implements the interpolation tour as explained in Young and Rheinagans: Visualizing Structure in High-Dimensional Multivariate Data. It centers the data at the origin and normalizes according to the standard deviation. The tour rotates through 90 degrees to the target plane and then through another 270 degrees back to the starting plane. The first argument X is the data matrix. T1 is a 2 element vector containing column indices for the first plane. Similarly, T2 is a 2 element vector containing indices for the target plane. These indices are for the principal components.

Input Arguments:

• X is the n by p data matrix.
• t1 is a 2-element vector containing column indices (i.e., the principal components) for the fist plane.
• t2 is a 2-element vector containing column indices (i.e., the principal components) for the second plane.

Output Arguments:

None.

Synopsis

t1 = [1,3]; % First and third PCs.

t2 = [2,4]; % Second and fourth PCs.

intour(X,t1,t2)

References:

Young, F. W., and Penny Rheingans. 1991. “Visualizing structure in high-dimensional multivariate data,” IBM Journal of Research and Development, 35:97-107.