The following question is from Roger W. Brockett, Finite Dimensional Linear Systems (Image of source):
Ordinarily one does not use division notation when dealing with matrices because $A/B$ might be interpreted as $AB^{-1}$ or $B^{-1}A$ and these are not necessarily the same. Why is the notation $(I+A)/(I-A)$ unambiguous?
I do not know what (I+A)/(I-A) is supposed mean, let alone proving it is unambiguous.
So the question is what that notation mean, and preferably why is unambigous?
Matrices $A$ and $B$ need not commute. So, $AB^{-1} \neq B^{-1}A$ in general. However, your text is claiming that matrices $(I+A)$ and $(I-A)^{-1}$ do commute for all matrices $A$ where $(I-A)$ is invertible. This can be proven thusly;
$$(I+A)(I-A)^{-1}=(I-A)^{-1}+A(I-A)^{-1}$$
$$\therefore (I-A)(I+A)(I-A)^{-1}=I+(I-A)A(I-A)^{-1} $$
$$\therefore(I-A)(I+A)(I-A)^{-1}=I+A(I-A)(I-A)^{-1}$$
$$\therefore(I-A)(I+A)(I-A)^{-1}=I+A$$
This yields
$$(I+A)(I-A)^{-1}=(I-A)^{-1}(I+A)$$
So the notation $(I+A)/(I-A)$ is unambiguous for matrices of this specific form.$$$$(Note: the division symbol assumes and requires that $(I-A)^{-1}$ exist)