Proving Invertibility of AB-I & BA-I When A & B Are Both nxn Matrices

  • Thread starter Thread starter daniel_i_l
  • Start date Start date
  • Tags Tags
    Matrices
Click For Summary

Homework Help Overview

The problem involves proving the invertibility of the matrix BA-I given that AB-I is invertible, with A and B being nxn matrices. The discussion explores the implications of matrix properties and relationships between determinants and invertibility.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to prove the statement by considering cases where either A or B is invertible, questioning the scenario when both are singular.
  • Some participants suggest examining the implications of non-invertibility and using proof by contradiction to explore the relationship between the matrices.

Discussion Status

The discussion is ongoing, with participants offering different lines of reasoning and approaches. There is no explicit consensus yet, but some productive directions have been suggested, particularly regarding the use of contradiction and vector implications.

Contextual Notes

Participants are navigating the complexities of matrix invertibility and determinants, with some expressing uncertainty about the implications when both matrices are singular.

daniel_i_l
Gold Member
Messages
864
Reaction score
0

Homework Statement


Q: If A and B are both nxn matrices and AB-I is invertable then prove that BA-I is also invertable.


Homework Equations


if A is invertible iff |A|<>0


The Attempt at a Solution


I've been thinking about this for over an hour I've only managed to prove it if either A or B are invertable. because if let's say A is invertable then:
|AB-I|<>0 => |AB-I||A|<>0 => |ABA-A|<>0 => |A||BA-I|<>0 => |BA-I|<>0 and so it's invertable. if B is invertable then you do pretty much the same thing on starting on the left side.
But what if they're both singular?
Thanks.
 
Physics news on Phys.org
X is not invertible if and only if there is a v=/=0 with Xv=0.

Suppose (AB-I)v=0, and see what you can deduce. (I don't promise this works, but is the first thing that springs to mind.)
 
You are too hung up on determinants. Try a proof by contradiction. Assume (BA-I) is NOT invertible. Then there is a nonzero vector x such that (BA-I)x=0. Now tell me what is (AB-I)Ax=?. (By playing the same game you did with the determinants).
 
Thanks a lot! now i can go to sleep...
 

Similar threads

  • · Replies 40 ·
2
Replies
40
Views
6K
Replies
4
Views
5K
  • · Replies 20 ·
Replies
20
Views
3K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 23 ·
Replies
23
Views
4K
  • · Replies 14 ·
Replies
14
Views
6K
  • · Replies 12 ·
Replies
12
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
7
Views
2K