Скачать или смотреть ITF lecture 2 continued:

The Inverse Trigonometric Functions
These notes amplify on the book’s treatment of inverse trigonometric
functions and supply some needed practice problems. Please see pages 543–
544 for the graphs of sin−1 x, cos−1 x, and tan−1 x.
1 The Arcsine Function
My sine is x; who am I? If x is any real number such that |x| ≤ 1, there are
infinitely many possible answers. For example, let x = 1/2. Then sin y = x
when y = π/6, 5π/6, −7π/6, −11π/6, 13π/6, 17π/6, −19π/6, and so on.
Note however that as y travels from −π/2 to π/2, sin y travels from −1 to
1. Since sin y is continuous and increasing in the interval −π/2 ≤ y ≤ π/2,
it follows that for any x between −1 and 1 there is exactly one y between
−π/2 and π/2 such that sin y = x. We can therefore define a new function
sin−1 as follows:
Definition 1. If −1 ≤ x ≤ 1, then sin−1 x (also known as arcsin x) is the
number between −π/2 and π/2 whose sine is equal to x.
Comment. If the function f has an inverse, that inverse is generally de-
noted by f −1. The notation sin−1 x that has just been introduced is (more
or less) in accord with this general convention. Not quite! The sine func-
tion does not have an inverse, since given sin t it is not possible to recover t
uniquely.
Maybe we are being too fussy. But the notation can also be a source
of confusion. For note that sin2 x is a standard abbreviation for (sin x)2,
and sin3 x is a standard abbreviation for (sin x)3. So should sin−1 x mean
(sin x)−1? Maybe it should. But it doesn’t!
The notation arcsin x is preferred by many mathematicians. Unfortu-
nately, sin−1 seems to be gaining ground over arcsin, maybe because it fits
the cramped space on calculator keyboards better. There are several other
notations, including “Arc sin,” and “asin.”
1
Pronunciations vary: sin−1 x can be pronounced “sine inverse (of) x,”
“inverse sine (of) x,” or even “arc sine x.”
Comment. You don’t have to know a lot about geography to know the
capital of the country whose capital is Amman. And you don’t have to
know much about trigonometry to find sin(sin−1(0.123)). Indeed it is clear
that
sin(sin−1 x) = x for all x between −1 and 1.
The behaviour of sin−1(sin x) is more complicated. Let x be any number
in the interval [−π/2, π/2]. Then the number between −π/2 and π/2 whose
sine is sin x is clearly x.
But suppose for example that π/2 ≤ x ≤ 3π/2. Since the sine function
is symmetrical about the line x = π/2, we have sin x = sin(π − x). And
since π − x lies between −π/2 and π/2,
sin−1(sin x) = sin−1(sin(π − x)) = π − x.
2 Differentiating the Arcsine Function
Let y = sin−1 x. Then for all x in the interval [−1, 1]
sin y = x.
Assume that y is differentiable at all x in (−1, 1)—it really is, but we omit
the proof. If we differentiate both sides of the equation above with respect
to x, then the Chain Rule gives
(cos y)
dy
dx = 1.
Thus if cos y 6= 0 then
dy
dx = 1
cos y = 1
cos(sin−1 x)
.
The above formula is correct but unattractive. To improve it, recall the
familiar identity
cos2 y + sin2 y = 1.
Since y = sin−1 x, we have
cos2 y = 1 − (sin y)
2 = 1 − (sin(sin−1 x))2 = 1 − x2.
2
But y lies in the interval (−π/2, π/2), and therefore cos y is positive. It
follows that cos y = √
1 − x2 and therefore
d
dx(sin−1 x) = 1
√
1 − x2 (1)
3 The Arctangent Function
The tangent function is continuous and increasing on the interval (−π/2, π/2).
Moreover, as t approaches −π/2 from the right, tan t becomes arbitrarily
large negative, while as t approaches π/2 from the left, tan t becomes ar-
bitrarily large positive. It follows that as t ranges over (−π/2, π/2), tan t
takes on every real value exactly once.
Thus the tangent function, restricted to the interval (−π/2, π/2), has an
inverse. We can therefore define a new function tan−1 as follows:
Definition 2. For any real number x, tan−1 x (also known as arctan x) is
the number between −π/2 and π/2 whose tangent is equal to x.
4 Differentiating the Arctangent Function
Let y = tan−1 x. Then for all x
tan y = x.
Assume that y is differentiable for all x—it really is, but we omit the proof.
If we differentiate both sides of the above equation with respect to x, we
obtain
(sec2 y)
dy
dx = 1.
It follows that
dy
dx = 1
sec2 y = 1
sec2(tan−1 x)
.
To make the above formula more attractive, recall the identity 1 + tan2 y =
sec2 y. If you do not recall it, you can quickly derive it by dividing both
sides of the identity cos2 y + sin2 y = 1 by cos2 y.
Since y = tan−1 x, we have
sec2 y = 1 + (tan y)
2 = 1 + (tan(tan−1 x))2 =1+ x2,
and therefore we can rewrite the formula for the derivative of tan−1 x as
d
dx(tan−1 x) = dy
dx = 1
1 + x2 (2)
3
5 The Arccosine Function
The cosine function decreases from 1 to −1 over the interval [0, π]. It is
therefore reasonable to define cos−1 x to be the number between 0 and π
whose cosine is x. By an argument almost identical to the argument in
Section 2, we can show that
d
dx(cos−1