We sometimes need to find an inverse of a function that is not one-to-one. We can do this by restricting the domain of the original function so that the new function with the restricted domain is one-to-one on the restricted domain. For example, the function
![\[f(x) = 2x^2 + 3\] \[f(x) = 2x^2 + 3\]](https://d11d2f3yu3bis.cloudfront.net/latex/d/3/6fbda9e7550f833a520f90e54988c7.gif)
is not one-to-one on its implied domain, the interval
![\[(-\infty, \infty).\] \[(-\infty, \infty).\]](https://d11d2f3yu3bis.cloudfront.net/latex/6/f/b6cda100041acf6809cc40abd45c12.gif)
However, if we restrict the domain to be the interval
![\[[0, \infty)\] \[[0, \infty)\]](https://d11d2f3yu3bis.cloudfront.net/latex/3/0/b6c278a72572c2f2aab47484c53ae1.gif)
then the new function is one-to-one.
Because the six trigonometric functions are all periodic, none of them is one-to-one (See
Topic 19.) In order to define an inverse for each of the trigonometric functions, we will restrict the domain so that the function is one-to-one on the restricted domain.
For the sine function, we want to restrict the domain to an interval over which the function passes the horizontal line test. Although we could choose other intervals, it is standard to choose the interval
![\[\left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\] \[\left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\]](https://d11d2f3yu3bis.cloudfront.net/latex/c/e/c3357c7ffe09299c680c7c5f50b56b.gif)
Therefore, we consider the function
![\[f(x) = \sin x, x \in \left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\] \[f(x) = \sin x, x \in \left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\]](https://d11d2f3yu3bis.cloudfront.net/latex/f/c/a6bf20f9f9f38edb98bac39fea3099.gif)
This restricted sine function is one-to-one and has an inverse, called the
inverse sine function and denoted
![\[\sin^{-1} x\] \[\sin^{-1} x\]](https://d11d2f3yu3bis.cloudfront.net/latex/7/5/43063bcc273fffab5721fb2ea8a1e4.gif)
or arcsin
x. The domain of the inverse sine function is
![\[[-1, 1]\] \[[-1, 1]\]](https://d11d2f3yu3bis.cloudfront.net/latex/6/c/36681b2a65c509153d11fa0331531f.gif)
and its range is
![\[\left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\] \[\left[- \frac{\pi}{2}, \frac{\pi}{2}\right].\]](https://d11d2f3yu3bis.cloudfront.net/latex/c/e/c3357c7ffe09299c680c7c5f50b56b.gif)
The graphs of
![\[y = \sin x, x \in \left[- \frac{\pi}{2}, \frac{\pi}{2}\right]\] \[y = \sin x, x \in \left[- \frac{\pi}{2}, \frac{\pi}{2}\right]\]](https://d11d2f3yu3bis.cloudfront.net/latex/a/d/cc2381834a698018a0102e66f79868.gif)
(the solid curve) and
![\[y = \sin^{-1} x, x \in [-1, 1]\] \[y = \sin^{-1} x, x \in [-1, 1]\]](https://d11d2f3yu3bis.cloudfront.net/latex/1/d/b67f67d0d9acade1bacceca05721aa.gif)
(the dashed curve) are shown.
For the tangent function, we again want to restrict the domain to an interval over which the function passes the horizontal line test. The standard interval is almost the same as the interval for the sine function, the interval
![\[\left(- \frac{\pi}{2}, \frac{\pi}{2}\right),\] \[\left(- \frac{\pi}{2}, \frac{\pi}{2}\right),\]](https://d11d2f3yu3bis.cloudfront.net/latex/d/6/aa0141471504236cbaf6f6978825dd.gif)
except the endpoints are not included because the tangent function is not defined at
![\[\frac{\pi}{2}\] \[\frac{\pi}{2}\]](https://d11d2f3yu3bis.cloudfront.net/latex/6/f/7a1fed28a67e8629dd7cf93b54351d.gif)
and
![\[- \frac{\pi}{2}.\] \[- \frac{\pi}{2}.\]](https://d11d2f3yu3bis.cloudfront.net/latex/5/3/f060739576892f31eeb1d0a4c7bb3d.gif)
Therefore, we consider the function
![\[f(x) = \tan x, x \in \left(- \frac{\pi}{2}, \frac{\pi}{2}\right).\] \[f(x) = \tan x, x \in \left(- \frac{\pi}{2}, \frac{\pi}{2}\right).\]](https://d11d2f3yu3bis.cloudfront.net/latex/d/8/d639c81de3307a3adc10e8dc83dadd.gif)
The restricted tangent function is one-to-one and has an inverse, called the
inverse tangent function and denoted
![\[\tan^{-1} x\] \[\tan^{-1} x\]](https://d11d2f3yu3bis.cloudfront.net/latex/2/2/ad64da7d55331756e8c6e9ce812a33.gif)
or
![\[\arctan x.\] \[\arctan x.\]](https://d11d2f3yu3bis.cloudfront.net/latex/0/1/82ce0e66c0e114bdc87fbadbe5dc3a.gif)
The domain of the inverse tangent function is
![\[(-\infty, \infty)\] \[(-\infty, \infty)\]](https://d11d2f3yu3bis.cloudfront.net/latex/5/c/e0819d3580309c17aa2b324d841f4d.gif)
and its range is
![\[\left(- \frac{\pi}{2}, \frac{\pi}{2}\right).\] \[\left(- \frac{\pi}{2}, \frac{\pi}{2}\right).\]](https://d11d2f3yu3bis.cloudfront.net/latex/1/7/71e1eb8ac13c0bdb6e632e7aa1cd44.gif)
The graphs of
![\[y = \tan x, x \in \left(- \frac{\pi}{2}, \frac{\pi}{2}\right)\] \[y = \tan x, x \in \left(- \frac{\pi}{2}, \frac{\pi}{2}\right)\]](https://d11d2f3yu3bis.cloudfront.net/latex/1/5/318c38f197e331600ec691b989a8dd.gif)
(the solid curve) and
![\[y = \tan^{-1} x, x \in (-\infty, \infty)\] \[y = \tan^{-1} x, x \in (-\infty, \infty)\]](https://d11d2f3yu3bis.cloudfront.net/latex/1/3/d55ee5a8b62371342d666e25a9234c.gif)
(the dashed curve) are shown.
The inverses of the other four trigonometric functions are defined in a similar way by restricting the domain of each to an appropriate interval and using the fact that, in general, the domain of
f = the range of
f −1 and the range of
f = the domain of
f −1.