18.

01 Calculus

Jason Starr Fall 2005

Lecture 30. December 6, 2005 Practice Problems. Course Reader: 6C­2. 1. Sequences By deﬁnition, a sequence of real numbers is a rule assigning to each counting number n an associated real number an . The integer n is called the index of the sequence. Usually the index begins with n = 1, but occasionally it begins with another integer (sometimes 0). Sequences are often speciﬁed by giving the ﬁrst few values, and letting the reader infer the rule, e.g., 1 1 1 a1 = , a 2 = , a 3 = , . . . 1 2 3 It is always better to give a precise deﬁnition of each sequence, e.g., an = 1 , n = 1, 2, . . . n

The most common notation for a sequence is (an )n≥1 .

aN +1 . The sequence an = L. . The number l is a lower bound for the sequence. then an = (−1)n diverges. 2.01 Calculus Jason Starr Fall 2005 A sequence (an )n≥1 converges to a limit L if the sequence becomes arbitrarily close to L. (iv) Let r be a real number. L + �). If r = −1. Tests for convergence/divergence. A sequence (an ) is bounded above if there exists a real number u such that for every index n. . there exists an integer N (depending on the sequence and �) such that for every integer n ≥ N . (bn )n≥1 and (cn )n≥1 be sequences such that for every index n. a sequence (an ) is non­increasing if for every index n. an+1 ≥ an . L = lim an . and if. . One useful test for convergence is the Squeezing Lemma. Let (an )n≥1 . . then an = 1 converges to 1. . There are 2 remaining cases. In other words. The squeezing lemma. even though it is bounded (it never gets bigger than 1 or smaller than −1). A sequence (an )n≥1 is called non­ decreasing if for every index n. (i) Let L be a ﬁxed real number. simply take � = |L1 − L2 |/2 in the deﬁnition above. 1. (iii) The sequence an = (−1)n diverges.18. 2. converges to 0 if |r| < 1 and diverges if |r| > 1. n = 0. In a precise sense. . Examples. n→∞ n→∞ then also (bn ) converges and its limit equals the limit of the other 2 sequences. . A sequence cannot have more than 1 limit: given 2 potential limits L1 and L2 . the tail of the sequence aN . A sequence which is either non­decreasing or non­increasing (but not both increasing and decreasing) is called monotone. A sequence (an ) is bounded below if there exists a real number l such that for every index n. the sequence (bn ) is “squeezed” between the sequences (an ) and (cn ). If (an ) and (cn ) converge. and stays arbitrarily close to L. an ≤ b n ≤ c n . If r = 1. 2. aN +2 . lim an = lim cn . |an − L| < �. an+2 ≤ an . In other words. The number u is an upper bound for the sequence. n→∞ A sequence which does not have a limit is said to diverge. converges to L. (ii) The sequence an = n diverges. Similarly. n = 1. More precisely. . this sequence “diverges to ∞”. an ≤ u. an ≥ l. are all numbers in the interval (L − �. Another test for convergence is the Monotone Convergence Test. A sequence which has a limit is said to converge. The sequence an = rn . . . and the limit is denoted by. the sequence converges to L if for every positive number �.

a non­increasing sequence converges if and only if it is bounded below and the limit is the greatest lower bound for the sequence. This will also be shown soon. The ﬁrst is the sequence of partial sums. The second is the sequence of partial absolute sums. deﬁned by. b n = a1 + a 2 + · · · + a n = n � k=1 ak . n ∞ � (−1)n n=1 n . (−1)n . If the sequence of partial sums (bn )n≥1 converges. Examples. sometimes it is said the series converges conditionally. there are 2 important related sequences. it is said the series k ak converges absolutely. Although it is not obvious. A non­decreasing sequence converges if and only if it is bounded above.18.100). 3. if the series converges absolutely.01 Calculus Jason Starr Fall 2005 Monotone Convergence Test. Similarly. then the series converges (this is a basic theorem from course 18. an = The alternating harmonic series. ∞ n � � ak := lim bn = lim ak . Series. so how could this fail for . the terms in the alternating harmonic series can be rearranged so that the sum converges to any real number you like! This sounds ridiculous: ﬁnite sums are independent of the order in which the summands are added. (Bn )n≥1 . The harmonic sequence is the sequence an = 1/n. the limit is called the series of (an )n≥1 . Since the sequence of partial absolute sums for the alternating sequence equals the sequence of partial sums for the harmonic sequence. and is denoted by. The alternating harmonic sequence is. the alternating harmonic series does not converge absolutely. It only converges conditionally. (bn )n≥1 . k=1 n→∞ n→∞ k=1 � In this case is is said the series � ak converges. the harmonic � series n 1/n diverges to ∞. As will be shown soon. Bn = |a1 | + |a2 | + · · · + |an | = n � k=1 |ak |. Given a sequence (an )n≥1 . does converge. If a series converges but does not converge absolutely. deﬁned by. In this case. the limit of the sequence is the least upper bound for the sequence. If the sequence of partial absolute sums (Bn )n≥1 k converges. As counter­intuitive as this might sound.

absolutely convergent series are much more useful than conditionally convergent series. Because the harmonic series n 1/n diverges. −1/3. Therefore. 106 . Let r be a real number and let (an )n≥0 be the geometric sequence. Thus the sequence (a� converges to 0. This gives the following. � 4. Similarly. denote by L the limit of the sequence of partial sums (bn ). Now add one more odd term. n ≥ 0. is a non­decreasing sequence. The pathology of the preceding paragraph occurs with any conditionally convergent series. Comparison Test. by the monotone convergence theorem. But because we add positive terms with a much higher frequency than negative terms. an = rn . say. To see this. we can make the partial sums converge to any value we like. Now add only the ﬁrst odd term −1/2. independent of the order in which terms are added. Let (an )n≥1 and (bn )n≥1 be sequences� such that for every index n. then the series n=1 an converges absolutely. it immediately follows that the series n=1 (−1) diverges (arguing the opposite is a favorite pasttime of “mathematical cranks”).01 Calculus Jason Starr Fall 2005 � inﬁnite sums? The answer is quite simple. the sequence of partial sums is diverging to +∞. then the series n an diverges. For ∞ n example. |an | ≤ |bn |. Contrapositively. it converges if and only if it is bounded above. Test for convergence/divergence of series. This has a negligible eﬀect. For every positive real number �. The most common technique for proving the sequence of partial absolute sums is bounded above is by comparing it to a larger series that is known to converge. Thus. But then for n ≥ N + 1. This is the� most basic test for divergence of a series. Now it is not so surprising that by adding the terms in a careful order. we could negative terms with a very high frequency and make the partial sums diverge to −∞. For this reason. if the sequence (an ) does not converge n) to 0. n=1 A number of common convergence tests in calculus textbooks come to nothing more than combining the comparison test with an analysis of the geometric series. using �/2 in the deﬁnition of convergence of (bn ).18. The sequence of partial absolute sums. there exists an integer N such that for every n ≥ N . It is a very important fact that every absolutely convergent series has only a single limit. eventually every term in the sequence contributes to one of the partial sums. The most basic test for absolute convergence of a sequence follows from the monotone convergence test. |bn − L| < �/2. the same � is true for 1/2n . � ∞ If the series ∞ bn converges absolutely. Continuing in this way. If a series n an converges. Bn = n � k=1 |ak |. . add it up a very large number of only the (positive) even terms in the alternating harmonic series to make the partial sum bigger than. then the sequence (an ) converges to 0. Now add a large number of the remaining even terms to make the partial sum bigger than 107 . |an | = |bn − bn−1 | = |(bn − L) − (bn−1 − L)| ≤ |bn − L| + |bn−1 − L| < �/2 + �/2 = �.

There are two tests that allow us to compare a given sequence (an )n≥∞ to a geometric sequence (rn )n≥1 . 1 converges. The ratio test. Then the sequence (an )n≥1 can be� compared to a sequence (Crn )n≥1 for some choice of C. The ﬁnal test uses improper integrals to get useful information about a series. if r = 0. the series ∞ rn converges to 1/(1 − r) if |r| < 1. Altogether. g(x) ≤ an for all n ≤ x ≤ n + 1. Then the partial sums are � bn = n + 1. Thus. 1 1 Bn = 1 + |r| + |r|2 + · · · + |r|n = + |r|n+1 . the geometric sequence rn+1 converges if and only if |r| < 1. Similarly. There is no information if the sequence converges to 1 or diverges. The only remaining case is when r = 1. lim n→∞ � an � exists. call it r. By high school algebra. Comparison to an improper integral. � � � an+1 � � � . This leads to the ratio test : The series ∞ an converges absolutely if the sequence n=1 |an+1 /an | converges to a real number r < 1 and diverges if the sequence |an+1 /an | converges to a real number r > 1 (in which case. call it r. which diverges to ∞. There is no information if the sequence converges to 1 or diverges. n=0 and diverges otherwise. 1 − |r| 1 − |r| �∞ n also converges. the partial sums are 1 − rn+1 1 1 n+1 − = r . Let (an )n≥1 be a sequence. if r �= 1. and diverges if |r| > 1 or r = −1. If the improper integral.01 Calculus Jason Starr Fall 2005 (by convention. � ∞ f (x)dx. � ∞ � ∞ g(x)dx. let g(x) ≥ 0 be a function on [1. which is essentially the geometric sequence. then the series n=1 an converges absolutely. ∞) such that for every integer n. In this case. f (x) ≥ an for all n ≤ x ≤ n + 1. Assuming r �= 1. exists. the sequence (an )n≥1 does not converge to 0). Then the sequence (an )n≥1 can be compared to a sequence (Crn )n≥1 . Let f (x) ≥ 0 be a function on [1. the ﬁrst term a0 is deﬁned to be 1). n→∞ lim � n |an |. the geometric series n=0 r converges absolutely to 1/(1 − r) if |r| < 1. This leads � � to the root test : The series ∞ an converges absolutely if the sequence n |an | converges to a real n=1 � number r < 1 and diverges if the sequence n |an | converges to a real number r > 1. On the other hand. 1 . ∞) such that for every integer n.18. If the improper integral. bn = 1 + r + · · · + r n = 1−r 1−r 1−r Observe this sequence depends on n only in the last term rn+1 . if the following limit. If the following limit. the sequence of partial absolute sums.

the series. 2. as n tends to ∞. n � � ∞ ∞ s s The series n=1 1/n equals 1 + n=2 1/n . Therefore the partial sums diverge. g(x) ≤ an = 1/n on the interval [n. n � The absolute partial sum of the series n ck is simply. The harmonic series. The partial sum of (cn ) is.18. � n n � 1 ck = dx = ln(n). For both directions. 1. Let (an )n≥1 be the harmonic sequence. 1 . Then for every integer n. Examples. n=1 � n+1 f (x)dx. or g(x)dx. f (x) ≥ 1/(n + 1)s for every x in [n. By the Fundamental Theorem of Calculus. Therefore the partial sums tend to 1/(s − 1). (n + 1)s Let f (x) be the function f (x) = 1/xs . which is the same as. n Let g(x) be the function g(x) = 1/x on the interval [1. or cn = g(x)dx. n + 1]. the harmonic series also diverges (very slowly). n + 1]. ns n=1 . 1 an = s . Example. the natural logarithms ln(n) also tend to ∞ (although very slowly – ln(n) does not get bigger than a ﬁxed real number R until n gets bigger than the much larger number eR ). also ns−1 tends to ∞. deﬁne the � n+1 cn = n diverges. By the comparison test. then the series sequence (cn ) by. k=1 1 1 The result follows. s s−1 � s−1 1−sx s−1 s−1n 1 x 1 Because s is bigger than 1. ∞ � 1 . by the comparison test. Let s > 1 be a real number. Then for each integer n. Deﬁne the sequence (an )n≥1 by. k=1 � n � n n � ck = f (x)dx. 1 x k=1 an = As n tends to ∞. The Riemann zeta function. 1+ ∞ � n=1 1 . Therefore. ∞). the partial sums of the sequence (cn ) are.01 Calculus Jason Starr Fall 2005 � ∞ an does not converge absolutely. �n � � n 1 1 1 � 1 � = 1 − 1 cn = dx = .

ns n=1 This function is of fundamental importance in number theory. . the Clay Mathematics Institute has oﬀered a \$1 million prize for an accepted. For one of these problems in particular. refereed solution. The ﬁrst couple are ζ(2) = π 2 /6 and ζ(4) = π 4 /90. There are very fundamental open problems about the Riemann zeta function. denoted ∞ � 1 ζ(s) := .18. It is also pops up in Fourier series and statistical mechanics. The value of this limit is called the Riemann zeta function at s. The values of ζ(s) when s is an even integer are known.01 Calculus Jason Starr Fall 2005 converges absolutely to a value bounded by 1/(s − 1).