Skip to main content

Recitation 9

Problem 1 Find limx0+xx1ln(x+1)\lim_{x \to 0^+}\frac{x^x-1}{\ln (x+1)}.

Solution The limit is of type 0/00/0. Moreover, we have limx0(xx1)(ln(x+1))=limx0xx(1+lnx)(1/(x+1))=limx0(1+x)xx(1+lnx)=.\lim_{x\to 0}\frac{(x^x - 1)'}{(\ln(x+1))'} = \lim_{x\to 0}\frac{x^x(1 + \ln x)}{(1/(x+1))} = \lim_{x\to 0}(1+x)x^x(1 + \ln x) = -\infty. By l’Hôpital’s rule, the limit is -\infty.

Problem 2 Find limx0(ex+excosx)1x2\lim_{x \to 0}(e^{x}+e^{-x}-\cos x)^{\frac{1}{x^2}}.

Solution The limit is of type 11^\infty. Let f(x)=ex+excosx1f(x) = e^x + e^{-x}-\cos x - 1. We have limx0(ex+excosx)1x2=limx0[(1+f(x))1f(x)]f(x)x2.\lim_{x \to 0}(e^{x}+e^{-x}-\cos x)^{\frac{1}{x^2}} = \lim_{x \to 0}\left[(1+f(x))^{\frac{1}{f(x)}}\right]^{\frac{f(x)}{x^2}}. Since limx0f(x)=0\lim_{x\to 0}f(x) = 0, limx0(1+f(x))1f(x)=e\lim_{x \to 0}(1+f(x))^{\frac{1}{f(x)}} = e. Moreover, limx0f(x)/x2\lim_{x\to 0}f(x)/x^2 is of type 0/00/0, and limx0f(x)(x2)=limx0exex+sinx2x\lim_{x\to 0}\frac{f'(x)}{(x^2)'} = \lim_{x\to 0}\frac{e^x - e^{-x} + \sin x}{2x} is again of type 0/00/0, and limx0f(x)(x2)=limx0ex+ex+cosx2=32.\lim_{x\to 0}\frac{f''(x)}{(x^2)''} = \lim_{x\to 0}\frac{e^x + e^{-x} + \cos x}{2} = \frac{3}{2}. By l’Hôpital’s rule, limx0f(x)x2=32\lim_{x\to 0}\frac{f(x)}{x^2} = \frac{3}{2} and the limit is e3/2e^{3/2}.

Problem 3 limx0+esin(1+cosx)sin(1+cosx)(ex1)2\lim_{x \to 0^+}\frac{e^{\sin (-1+\cos x)}\sin (-1+\cos x)}{(e^x-1)^2}.

Solution If limx0sin(1+cosx)(ex1)2\lim_{x \to 0}\frac{\sin (-1+\cos x)}{(e^x-1)^2} exists, then limx0esin(1+cosx)sin(1+cosx)(ex1)2=limx0esin(1+cosx)limx0+sin(1+cosx)(ex1)2=limx0sin(1+cosx)(ex1)2.\lim_{x \to 0}\frac{e^{\sin (-1+\cos x)}\sin (-1+\cos x)}{(e^x-1)^2} = \lim_{x\to 0}e^{\sin (-1+\cos x)}\lim_{x \rightarrow 0^+}\frac{\sin (-1+\cos x)}{(e^x-1)^2} = \lim_{x \to 0}\frac{\sin (-1+\cos x)}{(e^x-1)^2}. Note that limx0sin(1+cosx)(ex1)2\lim_{x \to 0}\frac{\sin (-1+\cos x)}{(e^x-1)^2} is of type 0/00/0 and

limx0(sin(1+cosx))((ex1)2)=limx0cos(1+cosx)(sinx)2(ex1)ex=limx0cos(1+cosx)2exlimx0sinxex1=12limx0sinxex1\begin{aligned}\lim_{x \to 0}\frac{(\sin (-1+\cos x))'}{((e^x-1)^2)'} = \lim_{x\to 0}\frac{\cos(-1+\cos x)(-\sin x)}{2(e^x - 1)e^x} \\ = \lim_{x\to 0}\frac{-\cos(-1+\cos x)}{2e^x}\lim_{x\to 0}\frac{\sin x}{e^x - 1} = -\frac{1}{2}\lim_{x\to 0}\frac{\sin x}{e^x - 1}\end{aligned} if limx0sinxex1\lim_{x\to 0}\frac{\sin x}{e^x - 1} exists.
Finally, limx0sinxex1=limx0sinxxxex1=1\lim_{x\to 0}\frac{\sin x}{e^x - 1} = \lim_{x\to 0}\frac{\sin x}{x}\frac{x}{e^x - 1} = 1. The answer is thus 12\frac{-1}{2}.

Problem 4 Use Taylor’s theorem to calculate the following limit: limx0cosx1+x22x44!x(sinxx+x33!)\lim_{x \rightarrow 0}\frac{\cos x- 1 + \frac{x^2}{2}-\frac{x^4}{4!}}{x(\sin x - x+\frac{x^3}{3!})}.

Solution The 6th order Taylor polynomial of cosx\cos x is 1x22!+x44!x66!.1 - \frac{x^2}{2!} + \frac{x^4}{4!} - \frac{x^6}{6!}. By Taylor’s theorem,

limx0cosx(1x22!+x44!x66!)x6=0,\lim_{x\to 0}\frac{\cos x - (1 - \frac{x^2}{2!} + \frac{x^4}{4!} - \frac{x^6}{6!})}{x^6} = 0, and so limx0cosx1+x22!x44!x6=16!.\lim_{x\to 0}\frac{\cos x - 1 + \frac{x^2}{2!} - \frac{x^4}{4!}}{x^6} = -\frac{1}{6!}. The 5th order Taylor polynomial of sinx\sin x is xx33!+x55!.x - \frac{x^3}{3!} + \frac{x^5}{5!}. By Taylor’s theorem, limx0sinx(xx33!+x55!)x5=0,\lim_{x\to 0}\frac{\sin x - (x - \frac{x^3}{3!} + \frac{x^5}{5!})}{x^5} = 0, and so limx0sinxx+x33!x5=15!.\lim_{x\to 0}\frac{\sin x - x + \frac{x^3}{3!}}{x^5} = \frac{1}{5!}. Finally, limx0cosx1+x22x44!x(sinxx+x33!)=limx0cosx1+x22!x44!x6limx0x5sinxx+x33!=16.\lim_{x \rightarrow 0}\frac{\cos x- 1 + \frac{x^2}{2}-\frac{x^4}{4!}}{x(\sin x - x+\frac{x^3}{3!})} = \lim_{x\to 0}\frac{\cos x - 1 + \frac{x^2}{2!} - \frac{x^4}{4!}}{x^6}\lim_{x\to 0}\frac{x^5}{\sin x - x + \frac{x^3}{3!}} = -\frac{1}{6}.

Problem 5 Use Taylor’s theorem to calculate 1.01\sqrt{1.01} with precision of at least 10610^{-6}.

Solution Let f(x)=1+xf(x) = \sqrt{1+x}. Note that f(x)=12(1+x)1/2,f(x)=14(1+x)3/2,f(x)=38(1+x)5/2.f'(x) = \frac{1}{2}(1+x)^{-1/2}, f''(x) = \frac{-1}{4}(1+x)^{-3/2}, f'''(x) = \frac{3}{8}(1+x)^{-5/2}. The 2nd order Taylor polynomial of 1+x\sqrt{1+x} is P(x)=f(0)+f(0)1!xf(0)2!x2=1+12x18x2.P(x) = f(0) + \frac{f'(0)}{1!}x - \frac{f''(0)}{2!}x^2 = 1 + \frac{1}{2}x - \frac{1}{8}x^2. By Taylor’s theorem, for some 0<c<0.010 < c < 0.01 f(0.01)P(0.01)=f(c)3!(0.01)3=(3/8)(1+c)5/23!106<116106<107.|f(0.01) - P(0.01)| = \frac{f'''(c)}{3!}(0.01)^3 = \frac{(3/8)(1+c)^{-5/2}}{3!}10^{-6} < \frac{1}{16}10^{-6} < 10^{-7}. Therefore f(0.01)f(0.01) is approximately P(0.01)P(0.01) up to an error of 10710^{-7}.

Problem 6 Use Taylor’s theorem to prove the following inequality, for every xRx \in \mathbb{R}: ex1+x+x2/2!+x3/3!+x4/4!+x5/5!.e^x \geq 1+x+x^2/2!+x^3/3!+x^4/4!+x^5/5!.

Solution The 5th order Taylor polynomial of exe^x is P(x)=1+x+x2/2!+x3/3!+x4/4!+x5/5!,P(x) = 1+x+x^2/2!+x^3/3!+x^4/4!+x^5/5!, and f(6)(x)=ex.f^{(6)}(x) = e^x. By Taylor’s theorem, for some cc between 00 and xx, exP(x)=ec6!x6,e^x - P(x) = \frac{e^c}{6!}x^6, which is always 0\ge 0.

Concavity

Definition If the graph of ff lies above (below) all of its tangents on an interval II, then it’s called concave upward (downward) on II.

Concavity test How ff'' helps determine the intervals of concavity? If f(x)>0f''(x) > 0 for all xx in II, then ff is concave upward on II. If f(x)<0f''(x) < 0 for all xx in II, then ff is concave downward on II.

Definition A point PP on curve y=f(x)y = f(x) is called an inflection point if ff is continuous there and curve changes from concave upward to concave downward, or from concave downward to concave upward, at PP.

Second derivative test Suppose ff'' is continuous near cc. If f(c)=0f'(c) = 0 and f(c)>0f''(c) > 0, then ff has a local minimum at cc. If f(c)=0f'(c) = 0 and f(c)<0f''(c) < 0, then ff has a local maximum at cc.

Example Let’s discuss the curve y=f(x)=x44x3y = f(x) = x^4 - 4x^3 with respect to concavity, points of inflection and local minima and local maxima. Note that f(x)=4x312x2=4x2(x3)f'(x) = 4x^3 - 12x^2 = 4x^2(x-3) and f(x)=12x224x=12x(x2)f''(x) = 12x^2 - 24x = 12x(x-2). The function is monotone decreasing on (,3)(-\infty, 3) and increasing on (3,)(3,\infty). Second derivative f(x)=0f''(x) = 0 gives x=0,2x = 0, 2. Thus (0,0)(0, 0) and (3,27)(3, -27) are inflection points and ff is concave upward on (,0)(-\infty, 0) and (2,)(2, \infty), and it is concave downward on (0,2)(0,2). First derivative f(x)=0f'(x) = 0 gives x=0,3x = 0, 3 are critical numbers. By the second derivative test, we know that x=3x = 3 is a local minimum. However x=0x = 0 is not local min/max because ff is monotone decreasing on (,3)(-\infty, 3).