Matrix Variables (MATRIX-END MATRIX command)

A matrix variable is created by a matrix statement that assigns a value to a variable name.

• A matrix variable name follows the same rules as those applicable to an ordinary IBM® SPSS® Statistics variable name.
• The names of matrix functions and procedures cannot be used as variable names within a matrix program. (In particular, the letter T cannot be used as a variable name because T is an alias for the TRANSPOS function.)
• The COMPUTE, READ, GET, MGET, and CALL statements create matrices. An index variable named on a LOOP statement creates a scalar with a value assigned to it.
• A variable name can be redefined within a matrix program without regard to the dimensions of the matrix it represents. The same name can represent scalars, vectors, and full matrices at different points in the matrix program.
• MATRIX-END MATRIX does not include any special processing for missing data. When reading a data matrix from a data file, you must therefore specify whether missing data are to be accepted as valid or excluded from the matrix.