Basic Matrix Algebra in SPSS

General Comments:

* * * * * *
Creating Matrices in SPSS

To write a matrix, enclose the elements of the matrix within braces ({}). Elements of each row are separated by commas. (The number of columns is implied by the number of elements in each row.) Rows are separated by semi-colons. Thus {2,4,3,7;1,5,3,1} is a 2 x 4 matrix that looks like

2 4 3 7

1 5 3 1

Matrices can be defined explicitly as above or read into the MATRIX processor using a READ or GET command. For example if the above matrix was created using a word processor and saved as an ASCII or TXT file with the name "MATRIX1.DAT" it can be read with the following command:

READ A /FILE=MATRIX1.DAT /FIELD = 1 to #.

Manipulating Matrices

COMPUTE C = {A,B}. Concatenate (Join) A and B vertically where A and B are matrices or vectors. (A & B must have the same number of rows.)

COMPUTE C = {A;B}. Concatenate (Join) A and B horizontally where A and B are matrices or vectors. (A & B must have the same number of columns.)

COMPUTE A = MAKE (#1, #2, #3). Create #1 by #2 rectangular matrix with value #3 in all elements.

COMPUTE I = IDENT (#1). Create a square identity matrix with #1 rows/columns.

COMPUTE A = MDIAG (#1) Create square matrix with values of vector V as main diagonal, elsewhere 0

COMPUTE NR = NROW (A). Finds # of rows in A.

COMPUTE NC = NCOL (A). Finds # of columns in A.

Key Matrix Operators

COMPUTE C = A + B. Add matrices A and B.

COMPUTE C = A * B. Multiply matrices A and B.

COMPUTE AT = T (A). Find the transpose of A.

COMPUTE INVA = INV (A). Calculate the inverse of A.

COMPUTE DETA = DET (A) Calculate the determinant of A.

COMPUTE DIAGA = DIAG (A). Create a column vector from the diagonal of matrix A.

COMPUTE EIGA = EVAL (A) Compute the eigenvalues of A and place in column vector.

COMPUTE RANKA = RANK (A). Assess the rank of A.

Results

PRINT A /FORMAT = F10.4 / TITLE = "My Results".