Maximum Dimension of K Matrix

Use this field to specify the maximum dimension of the K matrix to test for finding the optimal compression level of K. The value entered in this field must be a numeric positive integer.

If this field is left blank or if the value is greater than the dimension of the input K matrix, then the input K matrix (where each row is a sample, and no clustering has been done) is used in the optimization algorithm. In this case, the full K matrix can be chosen if it provides the optimal fit (hence no compression would have been done).

Note: If the trait is binary and PROC GLIMMIX is used, the minimum dimension to be tested is 2.

To Specify the Maximum Dimension of the K Matrix:

Make sure that the Optimized compression method is specified OR the Compress the K matrix box is checked (whichever is available).

Type the desired value in the text box.

If you change your mind and would like to reset the field, click Clear.