3.10. This problem considers the question, why not use UL instead of LU? You can assume that A is a 2 ×2 matrix. (a) Find a Doolittle version of the factorization A = UL, where L is a unit lower triangular matrix and U is upper triangular. You can assume pivoting is not necessary. (b) Describe the resulting algorithm for solving Ax = b. (c) Is there any connection between the Doolittle factorization of AT and the one you found in part (a)?

Respuesta :

Answer:

a

The Doolittle version of factorization A =UL is

                     [tex]\left[\begin{array}{ccc}a_{11}&a_{12\\a_{21}&a_{22}\end{array}\right] =\left[\begin{array}{ccc}a_{11}-a_{12}*\frac{a_{21}}{a_{22}} &a_{12}\\0&a_{22}\end{array}\right]\left[\begin{array}{ccc}1&0\\\frac{a_{21}}{a_{22}} &1\end{array}\right][/tex]

b

           Algorithm

1)  Obtain the values of the individual elements of U and L

            [tex]U_{22} = a_{22}[/tex]

            [tex]l_{21} = \frac{a_[21} {U_{22}}}[/tex]

            [tex]U_{12} = a_{12}[/tex]

            [tex]U_{11} = a_{11} - U_{12}*l_{21}[/tex]

2) Obtain the value of z for the expression [tex]Uz = b[/tex] 1.e

                       [tex]\left[\begin{array}{ccc}U_{11}&U_{12\\0&U_{22}\end{array}\right] \left[\begin{array}{ccc}z_1\\z_2\end{array}\right] =\left[\begin{array}{ccc}b_1\\b_2\end{array}\right][/tex]

 [tex]z_2 =\frac{b_2}{U_22}[/tex]

  [tex]z_1 = \frac{1}{U_[11}} [b_1 - U_{12}\ z_2][/tex]

 3   Obtain the value of x for the expression [tex]Lx = z[/tex] 1.e

              [tex]\left[\begin{array}{ccc}1&0\\l_{21}&1\end{array}\right] \left[\begin{array}{ccc}x_1\\x_2\end{array}\right] = \left[\begin{array}{ccc}z_1\\z_2\end{array}\right][/tex]

       [tex]x_1 = z_1[/tex]

        [tex]x_2 = z_2 -l_{21}x_1[/tex]

c

  Let say that  [tex]A^T =LU[/tex] is the Doolittle factorization of [tex]A^T[/tex]

where L is a unit lower triangular matrix and U is upper triangular matrix

     Hence [tex]A = U^T L^T[/tex]

Here  [tex]L^T[/tex] is a unit upper  triangular matrix and [tex]U^T[/tex] lower is triangular  

In the question that we are given  A = UL meaning that  [tex]A^T = L^TU^T[/tex] looking at this  U is upper triangular matrix and L is unit lower triangular matrix. So we can say that the Doolittle version given behave like Doolittle factorization only for A and there is no connection with [tex]A^T[/tex]

Step-by-step explanation:

From the question let assume that          [tex]A =\left[\begin{array}{ccc}a_{11}&a_{12\\a_{21}&a_{22}\end{array}\right][/tex]

The objective is to find the Doolittle version of the factorized  A = UL and this can be expressed in matrix as

                          [tex]\left[\begin{array}{ccc}a_{11}&a_{12\\a_{21}&a_{22}\end{array}\right] =\left[\begin{array}{ccc}U_{11}&U_{12\\0&U_{22}\end{array}\right]\left[\begin{array}{ccc}1&0\\l_{21}&1\end{array}\right][/tex]

Looking at this in form of matrix multiplication

                             [tex]a_{11} =U_{11} +U_{12} l_{21}[/tex]

                             [tex]a_{12} = U_{12}[/tex]

                            [tex]a_{21} = U_{22} l_{21}[/tex]

                            [tex]a_{22} = U_{22}[/tex]

           =>      [tex]l_{21} = \frac{a_{21}}{U_{22}} = \frac{a_{21}}{a_{22}}[/tex]

    and           [tex]u_{11} = a_{11} - U_{12}l_{21} = a_{11} -a_{12} *\frac{a_{21}}{a_{22}}[/tex]

   Hence            [tex]U = \left[\begin{array}{ccc}a_{11}-a_{12}*\frac{a_{21}}{a_{22}} &a_{12}\\0&a_{22}\end{array}\right][/tex]

                       

                          [tex]L = \left[\begin{array}{ccc}1&0\\\frac{a_{21}}{a_{22}} &1\end{array}\right][/tex]

 b)   From the question we are given

                 [tex]Ax =b[/tex]

And objective is to describe the algorithm to solve it

                [tex]Ax =b[/tex]

                [tex]UL x = b[/tex]

Let assume that [tex]z = Lx[/tex]

               So

                           [tex]U z = b[/tex]

   Given that  U is upper triangular  matrix then y can be obtained by back substitution

       Now that we are able to obtain y, given that  L is a unit lower triangular matrix we can obtain x by forward substitution

The algorithm is given in the answer section

ACCESS MORE