Question on solving equations by applying inverse functions

50 Views Asked by At

If we want to find solutions of the equation $x^2=1$, where $x\in\mathbb{R}$, I frequently read/see that you should simply apply the inverse function which delivers $x=\pm1$.


I am not sure if I am overthinking but I would like to know if the following thoughts are correct if you want to derive the solutions rigorously:

1.) Applying the inverse function to solve for $x$ only makes sense if the function is injective (on the associated set). However, applying $\sqrt{\cdot}$ to $x^2=1$ yields $\sqrt{x^2}=|x|=1$ and I am not sure what this means. Is $|x|=1$ a well defined expression? I guess it's not?!

2.) If we only admit $x\geq0$, then $x^2$ becomes injective and $+\sqrt{\cdot}$ is the inverse function. Applying it to $x^2=1$ yields $\sqrt{x^2}=|x|=x=\sqrt{1}=1$.

3.) If we only admit $x<0$, then $x^2$ becomes injective and $-\sqrt{\cdot}$ is the inverse function. Applying it to $x^2=1$ yields $-\sqrt{x^2}=-|x|=-(-x)=-\sqrt{1}=-1$. Hence, $-1$ is a solution.


Is this the rigourous idea behind saying that the soultions are $\pm1$?

PS: Maybe in the context of this simple equation it seems kind of obvious but if we generalize it to some equation $h(x)=1$ where $h:\mathbb{R}\to\mathbb{R}$ and $h$ is only injective on say $[0,1]$, then applying $h^{-1}$ to $h(x)=1$ might produce an expression $h^{-1}\circ h$ which doesn't look as nice as $|x|$. One can think for example of the equation $\sin(x)=\frac{1}{2}$ where $x\in\mathbb{R}$. So you have no choice but to first restrict the domain and then apply the inverse function.

1

There are 1 best solutions below

0
On BEST ANSWER

You are absolutely correct. In fact, when reading "applying the inverse function, we get $x=\pm1$" you should be confused by the fact that a function must return only one value, and in this case it would be returning two values ($1$ and $-1$).

Therefore, what is meant by that statement is, precisely, the whole procedure you just specified:

  • Considering restrictions that make the function bijective with its image
  • Applying each inverse function in the restrictions
  • Take the union of all those answers

Another idea to take into account is that, in order to be sure to get all the possible values that make that condition true ($x^2=1$) restrictions should cover the entire domain.