Can determinants be calculated using tensor notation?

  • Thread starter Thread starter Durato
  • Start date Start date
  • Tags Tags
    Tensor
Click For Summary

Homework Help Overview

The discussion revolves around the calculation of determinants using tensor notation, specifically focusing on the relationship between the determinants of two 3x3 matrices A and B and their product C = AB.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • The original poster attempts to express the determinant of the product of two matrices using indicial notation and seeks clarification on the connection between the determinant of the product and the product of the determinants.
  • Some participants suggest considering cases where one of the matrices is not invertible and discuss the implications for the determinant.
  • Others question whether the proof should strictly adhere to indicial notation as requested in the problem statement.
  • One participant provides an alternative approach using elementary matrices to express the determinants.

Discussion Status

The discussion is ongoing, with participants exploring different methods and interpretations. Some guidance has been offered regarding the use of indicial notation and the implications of matrix invertibility, but no consensus has been reached on a definitive approach.

Contextual Notes

Participants note the importance of adhering to the problem's requirements and the potential challenges posed by the use of indicial notation. There is also mention of a technical issue with the forum's post preview feature.

Durato
Messages
38
Reaction score
0
[SOLVED] Tensor notation-Determinants

I'm trying to learn the basics of Tensor calculus using a free online book (Introduction to Tensor Calculus and Continuum Mechanics), and I got stuck on this question (Part 2 in book, after non-math introduction).

link to part 2, questions (mine #19) at end:
http://www.math.odu.edu/~jhh/part2.PDF"

Homework Statement


Let A and B denote 3x3 matrices with elements A_{ij} and B_{ij}
respectively. Show that if C = AB is a matrix product, then det(C) = det(A)*det(B)
where det = determinant.

Hint: use the result from example 1.1-9

Homework Equations


det(A) = e_{ijk}A_{1i}A_{2j}A_{3k}

The Attempt at a Solution


The matrix multiplication of A*B =C, in indical notation, is
C_{ij}=A_{im}B_{mj} (I think) where the first subscript in A,B and C is the row number and the second subscript the column of the matrix.

Then, plugging into 'relevant equation' above, we get
det(C) = e_{ijk}C_{1i}C_{2j}C_{3k}
det(C) = e_{ijk}(A_{1m}B_{mi})(A_{2n}B_{nj})(A_{3x}B_{xk})

Then, I compare this to just multiplying det(A)*det(B)
det(A) * det(B) = (e_{ijk}A_{1i}A_{2j}A_{3k})(e_{rst}B_{1r}B_{2s}B_{3t})

However, from here, I can't seem to make the connection between the two:cry:
I would try to expand it, but I don't think this is the way to do it, since it would get rid of the advantage of indical notation. Help would be appreciated! Thxs!

PS: on a side note, whenever you preview a post, the template is pasted again to the end of the post, which is annoying. Maybe someone can fix this...
 
Last edited by a moderator:
Physics news on Phys.org
I believe you need to prove this in two cases. If B is not invertible, then det(B)=0. There is a non-trivial solution. Thus AB is not invertible.

Else, suppose A and B are invertible/nonsingular, then there are elementary matrices, A=Ej, E_j-1,... E_1
det(A)=det(E_j)*det(E_j-1)...*(E_1)

Now, det(AB)= det(E_j)*det(E_j-1)...(E_1) * det(B).
Then A and B can be expressed as a product of elementary matrices.

Edit:Since C=AB, then det(C)=det(AB). Hope that helps.
 
Last edited:
Maybe I'm reading the intent incorrectly, but I think the problem specifically wants him to prove it using indicial notation.
 
Oh sorry.

Let C=AB. By indicial notation, C_ij=A_ik B_kj

Since A is 3x3 matrix, then det(A)= A_11(A_22...) - A_12(A21..) + A_13(A_21...) by method of solving for determinants where n>2
Then det(A)= epsilon_ijk A_i1 A_j2 A_k3 = epsilon_ijk A_1i A_2j A_3k
Therefore, epsilon_rst det(A) = epsilon_ijk A_ir A_js A_kt

We know that det(C)=det(AB)=epsilon_ijk C_i1 C_j2 C_k3
=...= epsilon_rst B_r1 B_s2 B_t3 det(A)
 
I can't believe i didn't see it! Thanks guys!:smile:
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
14
Views
4K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 10 ·
Replies
10
Views
2K