ChiralSuperfields said:
Thank you for your reply
@fresh_42 !
Is my checking correct ##A \cdot A^{-1} \cdot B \cdot B^{-1} = I_A \cdot I_B = I##?
Many thanks!
Yes, but a bit short. The long version is:
\begin{align*}
(A\cdot B) \cdot (A\cdot B)^{-1}&=(A\cdot B) \cdot (B^{-1}\cdot A^{-1}) \\
&= A \cdot (B \cdot (B^{-1}\cdot A^{-1}))\\
&= A\cdot (( B\cdot B^{-1})\cdot A^{-1})\\
&= A\cdot (I\cdot A^{-1})\\
&= A \cdot A^{-1} \\
&= I
\end{align*}
This proves by using the associative law of multiplication that ##B^{-1}\cdot A^{-1}## is
a inverse of ##(AB)^{-1}.##
I leave it to you to show that there cannot be more than one inverse, making ##B^{-1}\cdot A^{-1}##
the inverse of ##(AB)^{-1}.## Same with the identity matrix. There can only be one so we do not need to distinguish between ##I_A## and ##I_B## or between left-identity ##I_L\cdot A=A## and right-identity ##A\cdot I_R=A.## Both are the same. This can also be proven.
These proofs are a bit like a puzzle playing around with the associative, possibly distributive law. A nice Sunday afternoon exercise. The trick is to proceed step by step and only use these laws plus the definitions, e.g. that ##I_L\cdot A=A## and ##A\cdot I_R=A.## Show that ##I_L=I_R\,!##