Ask Your Question
0

Strange result for Jordan_form()

asked 2023-01-27 23:13:42 +0100

Cyrille gravatar image
     M = matrix([[4,-1,6],[2,1,6],[2,-1,8]])
    MM=M.jordan_form(transformation=true)
    show(MM)
    show(MM[0])
    show(MM[1])
    show(MM[1]^(-1))
    show(MM[1]^(-1)*MM[1])
    show(MM[1]^(-1)*MM[0]*MM[1])
    show(bool(MM[1]^(-1)*MM[0]*MM[1]==M))

According to the documentation and what I know of Jordan decomposition the last line should be True. No ?

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted
0

answered 2023-01-28 03:44:18 +0100

achrzesz gravatar image

updated 2023-01-28 03:48:57 +0100

Jordan canonical form is given by J = T^{-1} A T

A = matrix([[4,-1,6],[2,1,6],[2,-1,8]])
J, T = A.jordan_form(transformation=True)
T^(-1)*(A * T) ==  J

True

You checked:

T^(-1)*J*T == A

which is not the same

edit flag offensive delete link more

Comments

Note that

T*J*T^(-1) == A

True
achrzesz gravatar imageachrzesz ( 2023-01-28 16:23:38 +0100 )edit

Ok but my formula is in the quickref-linalg.pdf . Someone should modify it.

Cyrille gravatar imageCyrille ( 2023-01-28 22:02:11 +0100 )edit

And I wonder what arr the matrix obtained by my way.

Cyrille gravatar imageCyrille ( 2023-01-29 04:54:21 +0100 )edit

The matrix P from quickref-linalg.pdf is inverse to the transformation T. Your MM[1] corresponds to P

achrzesz gravatar imageachrzesz ( 2023-01-29 06:48:52 +0100 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

1 follower

Stats

Asked: 2023-01-27 23:13:42 +0100

Seen: 133 times

Last updated: Jan 28 '23