Page 15 - 4263
P. 15

Rank of a matrix


                     Using basic transformations, any matrix can be brought to the view shown as below:




                                                                             
                                                    1    0    0   . . .  0  0
                                                    0    1    0   . . .  0  0
                                                                             
                                                                             
                                                   0    0    1   . . .  0  0  
                                            A =                                ,
                                                  . . . . . . . . . . . . . . . . . . 
                                                                             
                                                   0    0    0   . . .  1  0  
                                                    0    0    0   . . .  0  1



                 where on the “main diagonal” (this diagonal only in a square matrix leads to the right bottom
                 corner) are situated r units and all the rest of matrix elements are equal to zero. It’s obvious,
                 that the rank of such a matrix and therefore of the given matrix equals to r.
                     Let’s note that by means of basic transformations in practice it’s enough to reduce the matrix
                 to a triangular view if it’s square. If a given matrix is rectangular, it can be brought to a trapezium
                 view and the number of non-zero rows is equal to the rank of this matrix.

                     For example, let’s find the rank of such a matrix:



                                                                                              
                             2 −1 3 4 5             −1 2 3 4 5              1   −2 −3 −4 −5
                            4 −2 5 6 7           −2 4 5 6 7           −2    4    5    6    7  
                      A =                     ∼                    ∼                            ∼
                            6 −3 7 8 9           −3 6 7 8 9           −3    6    7    8    9  
                             2 −1 2 2 2             −1 2 2 2 2             −1    2    2    2    2
                                                           
                                       1 −2 −3 −4 −5
                                                                 (                      )
                                      0   0    1    2    3       1 −2 −3 −4 −5
                                  ∼                          ∼                           .
                                      0   0    2    4    6       0   0    1    2    3
                                       0   0    2    4    6




                 As we see, two non-zero rows are left. So, the rank of matrix A equals to 2.

                  Theorem 2.3.
                  If the rank of matrix A equals to r, then it is possible to find r of linear independent rows or
                  columns in this matrix. The other rows or columns can be linearly represented by previous
                  ones.                                                                                ⋆



                     It’s recommended to prove this theorem individually. We will only pinpoint the conse-
                 quences coming out of this theorem:

                    1. When transpositioning a matrix its rows become columns and its rank
                       doesn’t change. Therefore the maximum number of linearly indepen-
                       dent columns of a matrix equals to the maximum number of linearly
                       independent rows.
                    2. In order the determinant to be equal to zero, it is necessary and suffi-
                       ciently for its rows or columns to be linearly dependent. It means that at
                       least one row or column is represented by the rest of rows or columns.


                                                              15
   10   11   12   13   14   15   16   17   18   19   20