lOMoARcPSD|48850158

MA2108 Final Cheatsheet

Mathematical Analysis I (National University of Singapore)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Anthony Rrewn ([email protected])
 lOMoARcPSD|48850158

NUS MA2108 Finals Cheatsheet

Well-ordering principle for N Every non empty Applications of the Supremum Theorem 3.3.2 (Monotone Convergence the sequence (sn ) of partial sums is bounded. In this
subset S of N has a least element, i.e. there exists Theorem) Let (xn ) be a monotone sequence of real case,
m ∈ S such that m ≤ k, ∀k ∈ S.
Property numbers. Then
2.4.3 Archimedean Property If x ∈ R, then X
∃nx ∈ N : x < nx . xn = lim sn = sup{sn : n ∈ N}
The Real Numbers 1
Corollary 2.4.5 For any  > 0, ∃n ∈ N : n < .
(xn ) is convergent ⇐⇒ (xn ) is bounded. n7→∞

The Order Property of R Note that the numerator can be any constant c ∈ N. Moreover, Theorem 3.7.7 (Comparison Test) Let (xn ), (yn )
0
Let n := cn ∈ N. (a) If (xn ) is bounded and increasing, then be real sequences and suppose that for some K ∈ N,
Bernoulli’s Inequality If x > −1, then lim xn = sup{xn : n ∈ N} we have
(1 + x)n ≥ 1 + nx, ∀n ∈ N. c 1 (b) If (xn ) is bounded and decreasing, then
<  ⇐⇒ <
n0 n lim xn = inf{xn : n ∈ N}
Absolute Value and the Real Line Theorem 3.4.2 If (xn ) converges to x, then any
0≤xn ≤yn , ∀n ≥ K
Theorem 2.4.8 (The Density Theorem of Q) If
Theorem 2.2.2 (Properties of absolute value) subsequence (xnk ) also converges to x. That is,
x, y ∈ R with x < y, then there exists a rational Then
(a) |ab| = |a||b|, ∀a, b ∈ R number r ∈ Q : x < r < y. P
(b) |a|2 = a2 , ∀a ∈ R lim xnk = lim xnk = x. (a)PThe convergence of yn implies the convergence
Corollary (The Density Theorem of Irrational nk 7→∞ k7→∞ of xn .
(c) If c ≥ 0, then |a| ≤ c ⇐⇒ −c ≤ a ≤ c Numbers) If x, y ∈ R with x < y, then there exists P
Theorem. If every subsequence of (xn ) converges to (b)
P The divergence of xn implies the divergence of
(d) −|a| ≤ a ≤ |a|, ∀a ∈ R an irrational number z : x < z < y.
the same limit, then the limit of (xn ) is the same. yn (contrapositive of (a)).
Theorem 2.2.3 Triangle Inequality If a, b ∈ R,
then |a + b| ≤ |a| + |b| Intervals Theorem 3.4.5 If a sequence (xn ) has either of the Theorem 3.7.8 (Limit comparison Test) Suppose
following properties, then (xn ) is divergent. that (xn ), (yn ) are strictly positive sequences and
Corollary 2.2.4 For a, b ∈ R, one has Theorem 2.5.1 (Nested Interval Property) If
1. (xn ) has two convergent subsequences whose suppose that the following limits exists:
(i) ||a| − |b|| ≤ |a − b| ≡ −|a − b| ≤ |a| − |b| ≤ |a − b| In = [an , bn ], n ∈ N is a nested sequence of closed
bounded intervals, then there exists a number limits are not equal.
(ii) |a − b| ≤ |a| + |b| xn
ξ ∈ R : ξ ∈ In , ∀n ∈ N. 2. (xn ) is unbounded. r := lim ( )
Corollary 2.2.5 If a1 , a2 , · · · , an are any real n7→∞ yn
Theorem 2.5.2 If In = [an , bn ], n ∈ N is a nested Theorem 3.4.8 (Bolzano-Weierstrass Theorem)
numbers, then
sequence of closed bounded intervals such that the Every bounded sequence has a convergent P P
lengths bn − an of In satisfy subsequence. (a) If r > 0 then xn is convergent iff yn is
Definition 3.5.1 (Cauchy Sequence) convergent.
|a1 + a2 + · · · + an | ≤ |a1 | + |a2 | + · · · + |an | ∀ > 0, ∃H() ∈ N : |xn − xm | < , ∀n, m ≥ H(). For
P P
inf{bn − an : n ∈ N} = 0, (b) If r = 0 and if yn is convergent, then xn is
the negation of 3.5.1, note that there can be a convergent.
then the number ξ contained in all In is unique. relation between n, m. Remark
Definition 2.2.7 Let a ∈ R and  > 0. Then the
Theorem 3.5.5 (Cauchy Convergence
−neighborhood of a is the set (i) The comparison tests 3.7.7 and 3.7.8 depend on
V (a) := {x ∈ R : |x − a| < }.
Sequences and their limits Criterion) A sequence of real numbers is convergent
having a stock of series that one knows to be
Definition 3.1.3 (Definition of Limits) iff it is a Cauchy sequence.
convergent (or divergent). The reader will find that
∀ > 0, ∃K() ∈ N : |xn − x| < , ∀n ≥ K(). Definition 3.5.7 A sequence (xn ) is called
the p-series is often useful for this purpose.
Theorem 3.1.9. Let X = (xn : n ∈ N) be a sequence contractive if there exists a constant C, 0 < C < 1
such that (ii) Contrapositive argument for divergence test.
of real numbers and let m ∈ N. Then the m-tail
Xm = (xm+n : n ∈ N) of X converges iff X
converges. In this case, lim Xm = lim X. |xn+2 − xn+1 | ≤ C|xn+1 − xn | Examples for Infinite Series
P∞ 1
Definition 3.2.1 A sequence X = (xn ) of real Example 3.7.6(b) The harmonic series n=1 n
numbers is said to be bounded if there exists a real ∀n ∈ N. The number C is called the constant of the diverges.
The Completeness Property of R number M > 0 : |xn | ≤ M, ∀n ∈ N. contractive sequence.
Example 3.7.6(c) The 2-series
P∞ 1
Theorem 3.5.8 Every contractive sequence is n=1 n2
Definition 2.3.2(a) Let S be a nonempty subset of Theorem 3.2.2 A convergent sequence of real converges.
R. A number u is said to be supremum (or least numbers is bounded. However, it is not true that Cauchy and so is convergent. P∞ 1
Example 3.7.6(d) The p-series n=1 np
upper bound) of S if it satisfies the conditions bounded sequence implies convergence.
(1) u is an upper bound of S (i.e. s ≤ u, ∀s ∈ S). By Theorem 3.2.3. If lim xn = x, lim yn = y and
Examples for Sequences and their converges when p > 1.
1
P∞
Example 3.7.6(e) The p-series n=1 np diverges
defn of upper bound, u ∈ R.
c ∈ R, then
n7→∞ n7→∞
limits. when 0 < p ≤ 1.
(2) if v is any upper bound of S, then u ≤ v. 1
(i) lim (xn + yn ) = x + y; Example 3.1.6(a) lim( n ) = 0.
n7→∞
Example 3.7.6(f ) The alternating harmonic series
Definition 2.3.2(b) Let S be a nonempty subset of Example 3.1.6(b) lim( n21+1 ) = 0. given by
R. A number w is said to be infimum (or greatest (ii) lim (xn − yn ) = x − y; √ √
n7→∞
lower bound) of S if it satisfies the conditions: Example 3.1.6(d) lim( n + 1 − n) = 0.
(iii) lim (xn yn ) = xy; Example 3.1.6(e) If 0 < b < 1, then lim(bn ) = 0.
n7→∞
(1) w is an lower bound of S (i.e. w ≤ s, ∀s ∈ S). By
defn of lower bound, w ∈ R.
(iv) lim c(xn ) = cx;
n7→∞
Example 3.1.11(c) c > 0 =⇒ lim(c1/n ) = 1. X∞
(−1)n+1 1 1 1 (−1)n+1
Example 3.1.11(d) lim(n1/n ) = 1. = − + − ··· + + ···
(2) if t is any lower bound of S, then t ≤ w. (v) lim (xn /yn ) = x/y provided yn 6= 0, ∀n ∈ N and n 1 2 3 n
n7→∞ n=1
Example 3.2.8 (f ) lim( sin n
n ) = 0.
y 6= 0. √
3
Observations 18/19 Midterm Q3(ii) lim( √ 3n ) = 1.
is convergent to ln 2.
Let k ∈ N. lim (ak k
n ) = ( lim (an )) .
n +n
P∞ 1
n7→∞ n7→∞ T7Q2(b) n=0 n! = en = e. The series
T5Q2 If xn > 0 and lim xn = x 6= 0, then Introduction to Infinite Series n7→∞
n7→∞ converges.
lim (xn )r = xr for r ∈ Q. Theorem 3.7.3 (the n-th term test) If the series P∞ 1 n
n7→∞
P
xn converges, then lim xn = 0. Or T7Q2(c) n=0 (1 + n) converges.
n7→∞
Theorem 3.2.7 (Squeeze Theorem) If
contrapositively, if lim xn does not exist or exists
xn ≤ yn ≤ zn , ∀n ∈ N and lim xn = lim zn = a,
n7→∞ n7→∞
n7→∞
but not 0, then the series diverges.
Infinite Series
then we have
Theorem
P 3.7.4 (Cauchy criterion test) The series Absolute Convergence
xn converges if and only if for every  > 0, there
lim yn = a
P
exists M () ∈ N such that if m > n ≥ M (), then Definition 9.1.1 The series xn is absolutely
n7→∞
|xn | is convergent. A
P
convergent if the series
Theorem 3.2.9 Let the sequence X = (xn ) converge |sm − sn | = |xn+1 + · · · + xm | <  series is said to be conditionally (or
to x. If x = lim(xn ), then |x| = lim(|xn |). nonabsolutely) convergent if it is convergent, but
The Supremum/Infimum Property (Axiom) of HW2Q1(c) Let (xn ) be a sequence of positive real Theorem 3.7.5 (Partial sum bounded test for it is not absolutely convergent.
P∞
R Every nonempty subset of R that has an numbers such that L := lim(xn+1 /xn ) exists. If series with nonnegative terms PSuppose Theorem 9.1.2 If a series n=1 xn is absolutely
upperbound/lowerbound has a supremum/infimum. Downloaded
|L| < 1, then (xn ) converges by n
and lim(x Anthony
) = 0. Rrewn ([email protected])
xn ≥ 0, ∀n ∈ N. Then the series xn converges iff convergent, then it is convergent.
 lOMoARcPSD|48850158

Tests for Absolute Convergence Theorem 4.2.4 Suppose that lim f (x) = L and
x7→c
Continuous Functions
Theorem 9.2.1 (Limit Comparison Test II) g(x) = M . Let b ∈ R. Continuous Functions
Suppose that (xn ), (yn ) are non-zero sequences and x7→c
suppose that the following limits exists: (a) lim (f ± g)(x) = L ± M ; Definition 5.1.1 ( − δ definition of continuity)
x7→c Let A ⊂ R, let f : A 7→ R and let c ∈ A. We say that
(b) lim (f g)(x) = LM, lim (bf )(x) = bL; f is continuous at c if given any number  > 0,
|xn | x7→c x7→c
r := lim ( ) there exists δ > 0 such that if x is any point of A
n7→∞ |yn | (c) If h(x) 6= 0, ∀x ∈ A and lim h(x) = H 6= 0, then
x7→c satisfying |x − c| < δ, then |f (x) − f (c)| < .
Equivalent Definition to 5.1.1. If c is a cluster
f L
P
(a)
P If r > 0, then xn is absolutely convergent iff point in A, f (x) is continuous at c iff
lim ( )(x) = .
yn is absolutely convergent. x7→c h H
P
(b)
P If r = 0 and yn is absolutely convergent, then f (c) = lim f (x)
xn is absolutely convergent. Remark Let f1 , · · · , fn be functions on A to R. x7→c

Theorem 9.2.2 (Root Test) Let (xn ) be a sequence Assume that lim fi (x) = Li , 1 ≤ i ≤ n.
x7→c Remark. The equivalent definition is useful as it
(a) If there exist r ∈ R with 0 ≤ r < 1 and K ∈ N Remark 1. lim (f1 + · · · + fn )(x) = L1 + · · · + Ln . opens up limit theorems (e.g. squeeze theorem, lower
such that x7→c
1. The cluster point c does not necessarily belong to bound etc).
2. lim (f1 · · · · · fn )(x) = L1 · · · · · Ln .
1
A. Thus to discuss the limit of f at a point x = c, we x7→c
|xn | n ≤ r ∀n ≥ K do not require f to be defined at x = c. 3. If L = lim f and n ∈ N, then
x7→c
2. Even if f (c) is defined, its value has no bearing on lim (f (x)n ) = ( lim f (x))n = Ln .
lim f (x). x7→c x7→c
P
then the series xn is absolutely convergent. x7→c
(b) If there exists K ∈ N such that Theorem 4.2.7 (Squeeze Theorem) Let A ⊆ R,
Definition If f has no limit at x = c, then we say
let f, g, h : A 7→ R and let c ∈ R be a cluster point of
that f diverges at c.
1 A. Suppose that f (x) ≤ g(x) ≤ h(x) for all x ∈ A
|xn | n ≥ 1 ∀n ≥ K Theorem 4.1.8 (Sequential Criterion for Limits
and lim f (x) = lim h(x) = L, then
of Functions) Let f : A 7→ R and a be a cluster x7→c x7→c

then the series

P
xn is divergent. point of A. The following statements are equivalent:
1. lim f (x) = L lim g(x) = L. Remarks
Corollary 9.2.3 Suppose that the limit x7→a x7→c
(i) Limit has to exist (i.e. left and right limit must
1
r := lim |xn | n exists. Then
P
xn is absolutely 2. For every sequence (xn ) in A that converges to a exist and be the same.)
n7→∞ such that xn 6= a for all n, the sequence (f (xn )) Theorem 4.2.9 (Lower Bound) If lim f (x) > 0 (ii) Equality must hold
convergent when r < 1 and is divergent when r > 1. x7→c
converges to L. (resp. lim f (x) < 0), then there exists Vδ (c) of c such (iii) f (c) must be well defined.
Remark. (i) If r = 1, the root test provides no x7→c
information. that f (x) > 0 (resp. f (x) < 0) for all Theorem 5.1.3 (Sequential Criterion for Continuity)
x ∈ A ∩ Vδ (c), x 6= c. A function f : A 7→ R is continuous at the point
(ii) Corollary 9.2.3 uses lim while Theorem 9.2.2 uses
c ∈ A iff for every sequence (xn ) in A that converges
upper/lower bound. Remark. Lower bound theorem can be applied to
to c, the sequence (f (xn )) converges to f (c).
Theorem 9.2.4 (Ratio Test) Let (xn ) be a continuous points as well since lim f (x) = f (c) iff f
n7→∞
sequence of nonzero real numbers. is continuous at c.
(a) If there exist r with 0 < r < 1 and K ∈ N such Definition 4.3.1 (One-sided limit) Let f be a
that function on A to R.
xn+1 (i) Let c be a cluster point of A ∩ (c, ∞). We say that
| |≤r ∀n ≥ K L is the right-hand limit of f at c if for any  > 0,
xn ∃δ > 0 :
P
then xn is absolutely convergent.
(b) If there exists K ∈ N such that
0 < x − c < δ(i.e.x ∈ (c, c + δ)) =⇒ |f (x) − L| < 
xn+1
Limit Theorems
| |≥1 ∀n ≥ K Definition 4.2.1 Let f : A 7→ R and c be a cluster
xn In this case, we write lim f (x) = L.
point of A. We say that f is bounded on a
x7→c+
P neighbourhood of c if there exists Vδ (c) and a
then xn is divergent. (ii) Let c be a cluster point of A ∩ (−∞, c). We say
constant M > 0 such that |f (x)| ≤ M for all
Corollary 9.2.5 Let (xn ) be a sequence of nonzero x ∈ A ∩ Vδ (c). that L is the left-hand limit of f at c if for any
real numbers and suppose that the limit  > 0, ∃δ > 0 :
x Theorem 4.2.2 If f : A 7→ R has a limit at a cluster
r := lim | n+1
xn | exists. Then
P
xn is absolutely point c, then f is bounded on some neighbourhood of
n7→∞
convergent when r < 1 and is divergent when r > 1. c.
Definition 4.2.3 Let A ⊂ R and let f and g be −δ < x − c < 0(i.e.x ∈ (c − δ, c)) =⇒ |f (x) − L| <  Theorem 5.1.4 (Discontinuity Criterion) f is
functions defined on A. We define the sum f + g, the
Limits difference f − g, and the product f g on A to be
discontinuous at x = a iff there exists a sequence
(xn ) in the domain of f such that xn 7→ a, but
the functions given by In this case, we write lim f (x) = L. f (xn ) 9 f (a).
Limits of Functions x7→c−
Theorem 4.3.2 (Sequential Criterion for Remark (i) (f (xn )) is divergent or (ii)
Definition 4.1.1 Let A be a subset of R. A point c (f + g)(x) := f (x) + g(x)
one-sided limits). lim f (xn ) 6= f (a).
is called a cluster point of A if for every δ > 0, n7→∞
there exists at least one point x ∈ A, x 6= c such that (f − g)(x) := f (x) − g(x) (i) lim f (x) = L ⇐⇒ if for every (xn ) converges
|x − c| < δ or c − δ < x < c + δ. Alternatively, (f g)(x) := f (x)g(x) x7→c+ Combinations of Continuous
to c and xn > c for all n ∈ N, the sequence (f (xn ))
converges to L.
Functions
(Vδ (c)\{c}) ∩ A 6= ∅ for any δ > 0 If b ∈ R, we define the multiple bf to be the
(ii) lim f (x) = L ⇐⇒ for every (xn ) converges to Theorem 5.2.1 Suppose that f and g are
function given by
x7→c− continuous at x = c, then
Definition 4.1.4 Let A ⊆ R and c be a cluster point c and xn < c for all n ∈ N, the sequence (f (xn )) (a) f ± g, f · g, and bf are also continuous at x = c,
of A. For a function f : A 7→ R, a real number L is (bf )(x) := bf (x) converges to L. where b is a constant.
said to be a limit of f at c if for any given  > 0, Theorem 4.3.3 lim f (x) = L exists iff both (b) If g(c) 6= 0, then f /g is also continuous at x = c.
If h(x) 6= 0, ∀x ∈ A, we define the quotient f /h to x7→c
there exists a δ = δ() > 0 such that if x ∈ A and Theorem 5.2.2 Suppose that f and g are
be the function given by lim f (x) and lim f (x) exist and
0 < |x − c| < δ, then |f (x) − L| < , that is, x7→c+ x7→c− continuous on A, then
f f (x) (a) f ± g, f · g, and bf are also continuous on A.
x ∈ A ∩ (Vδ (c)\{c}) =⇒ f (x) ∈ V (L) ( Downloaded
)(x) := by Anthony Rrewn ([email protected])
lim f (x) = lim f (x) = L. (b) If g(c) 6= 0, then f /g is also continuous on A.
h h(x) x7→c+ x7→c−
 lOMoARcPSD|48850158

Theorem 5.2.6 Let f : A 7→ R, g : B 7→ R and Theorem 5.3.10 (Preservation of Closed Monotone and Inverse Functions A Glimpse into Topology
f (A) ⊆ B. If f is continuous at c, and g is Intervals Theorem) If f is continuous on [a, b], Theorem 5.6.1 (One-sided Limits for Monotone
continuous at b = f (c), then g ◦ f is continuous at c. then Functions Exist Theorem) Let I ⊂ R be an
Theorem 5.2.7 Let f : A 7→ R, g : B 7→ R and interval and let f : I 7→ R be increasing on I. Open and Closed Sets in R
f (A) ⊆ B. If f is continuous at A, and g is Suppose that c ∈ I is not an endpoint of I. Then
continuous at B, then g ◦ f is continuous on A. f ([a, b]) := {f (x) : x ∈ [a, b]} = [m, M ] (i) lim f (x) = sup{f (x) : x ∈ I, x < c} Definition 11.1.1 A set V is called a
x7→c−
neighbourhood of a point x ∈ R if
Continuous Functions on Intervals where m = inf f ([a, b]) = f (x∗ ) and
(ii) lim f (x) = inf{f (x) : x ∈ I, x > c}
∃ > 0 : V (x) ⊆ V .
x7→c+
Definition 5.3.1 A function f : A 7→ R is bounded M = sup f ([a, b]) = f (x∗ ). That is, for any Remark. By showing limit exists, we can also use
on A if there exists M > 0 such that m ≤ k ≤ M , there exists c ∈ [a, b] such that f (c) = k. theorems related to limits such as the sequential Definition 11.1.2
Remark. Let f be a continuous function on [a, b]. criterion for limits theorem.
|f (x)| ≤ M, ∀x ∈ A Definition If f : I 7→ R is increasing on I and if c is (i) A subset G of R is open in R if G is a
1. f ([a, b]) = [m, M ] does not imply not an endpoint of I, we define the jump of f at c neighbourhood of any point in G, that is,
Or, the set f (A) is bounded. f ([a, b]) = [f (a), f (b)]. Image of the closed interval ∀x ∈ G, ∃x > 0 : Vx (x) ⊆ G.
Definition 5.3.3 (i) We say that f has an absolute to be
might not equal the closed interval given by
maximum on A if there exists x∗ ∈ A such that f (a), f (b). (ii) A subset F of R is closed in R if the complement
jf (c) := lim f (x) − lim f (x) C(F ) := R\F is open in R. Equivalently, for any
∗
f (x ) ≥ f (x), ∀x ∈ A. 2. If we replace the closed bounded interval [a, b] by x7→c+ x7→c− y∈ / F, ∃0 > 0 : V0 (y) ∩ F = ∅.
an arbitrary interval I (e.g. open or half-open), then
So in this case, f (x∗ ) = sup f (A) = max f (A). f (I) could be of any type such as half-open or = inf{f (x) : x ∈ I, x > c} − sup{f (x) : x ∈ I, x < c}
Theorem 11.1.4 (Open Set Properties)
(ii) We say that f has an absolute minimum on A unbounded.
if there exists x∗ ∈ A such that At the endpoints a or b, define
(a) The union of an arbitrary collection of open
Uniform Continuity subsets in R is open.

f (x∗ ) ≤ f (x), ∀x ∈ A.  lim f (x) − f (a) if c = a
x→a+
jf (c) :=
Uniform Continuity Let A ⊂ R, f : A 7→ R. We say  f (b) − lim f (x) if c = b (b) The intersection of any finite collection of open
So in this case, f (x∗ ) = inf f (A) = min f (A). that f is uniformly continuous on A if for each x→b−
Theorem 5.3.4 (Maximum-Minimum Theorem) sets in R is open.
 > 0, there exists a δ() > 0 (δ() only depends on ) Theorem 5.6.3 Let f : I 7→ R be increasing on I.
If f is continuous on [a, b], then f has an absolute such that Then f is continuous at c iff jf (c) = 0. Theorem 11.1.5 (Closed Set Properties)
maximum and an absolute minimum on [a, b].
Remark. Theorem 5.6.1, Corollary 5.6.2, defn of
Bisection Method. Essentially Binary Search.
Jump, and theorem 5.6.3 can be reformulated from (a) The intersection of an arbitrary collection of
Theorem 5.3.5 (Location of Roots Theorem) If
increasing to decreasing functions. closed subsets in R is closed
f is continuous on [a, b] and f (a)f (b) < 0, then there ∀x, y ∈ A, |x − y| < δ() =⇒ |f (x) − f (y)| <  Theorem 5.6.4 (Discontinuous points of
exists a point c ∈ (a, b) such that f (c) = 0. (b) The union of any finite collection of closed sets in
monotone functions) Let I ⊆ R be an interval and
let f : I 7→ R be monotone on I. Then the set of R is closed.
Negative Definition. f is not uniformly points D ⊆ I at which f is discontinuous is a
continuous on A if there exists an 0 > 0 such that countable set. Theorem 11.1.7 (Characterisation of Closed
for all δ > 0, there are points xδ , yδ such that Theorem 5.6.5 (Continuous Inverse Theorem) Sets) A subset F of R is closed iff for any convergent
|xδ − yδ | < δ and |f (xδ ) − f (yδ )| ≥ 0 . Let I ⊂ R be an interval and f : I 7→ R be strictly sequence (xn ) in F , lim xn belongs to F .
n7→∞
(Sequential Criterion for Uniform Continuity) monotone and continuous. Then the inverse function
The function f : A 7→ R is uniformly continuous on A f −1 is also strictly monotone and continuous on Theorem 11.1.8 A subset F of R is closed iff it
iff for any two sequences (xn ), (yn ) in A such that J := f (I) = R(f ). contains all of its cluster points.
lim xn − yn = 0, we have lim f (xn ) − f (yn ) = 0. Remark. f −1 : J 7→ R, where J might not equal the
n7→∞ n7→∞
codomain of f . Theorem 11.1.9 (Characterisation of Open
Negative Definition. There exists an 0 > 0 and Rational Power Function. xr , r ∈ Q is defined. Sets) A subset of R is open iff it is the union of
two sequences (xn ), (yn ) in A such that countably many disjoint open intervals in R.
lim xn − yn = 0 and we have
Theorem 5.3.7 (Intermediate Value Theorem) n7→∞
Let I be an interval, f be continuous on I, and lim |f (xn ) − f (yn )| ≥ 0 .
n7→∞
a, b ∈ I with f (a) ≤ f (b). For any k ∈ [f (a), f (b)],
Theorem 5.4.3 (Uniform Continuity Theorem)
Continuous Functions
then there exists a point c in I such that f (c) = k.
Remark. a, b ∈ I. (i) Case a ≤ b: [a, b] ⊆ I (ii) Case If f is continuous on a closed bounded interval [a, b],
then it is uniformly continuous on [a, b]. Lemma 11.3.1 A function f : A 7→ R is continuous
b > a: [b, a] ⊆ I.
at the point c in A iff for every (open) neighborhood
Definition 5.4.4 A function f : A 7→ R is said to be U of f (c), there exists a (open) neighborhood V of c
a Lipschitz function (or satisfy a Lipschitz such that if x ∈ V ∩ A, then f (x) ∈ U . That is, ∀
condition) on A if there exists a K > 0 such that neighborhood U of f (c), ∃ a neighborhood V of c st.
f (V ∩ A) ⊆ U .
|f (x) − f (y)|
≤ K, for x 6= y ∈ A Theorem 11.3.2 (Global Continuity Theorem)
|x − y| Let f : A 7→ R be a function with domain A. Then
the following are equivalent:
which implies that its derivative f 0 (x) (if exists) is
bounded on A. (a) f is continuous at every point of A.
Theorem 5.4.5 If f : A 7→ R is a Lipschitz function, (b) For every open set G in R, there exists an open
then f is uniformly continuous on A. set H in R such that H ∩ A = f −1 (G).
Theorem 5.4.7 (Uniformly continuous functions
preserve Cauchy sequence) If f : A 7→ R is Note that for a set G, f −1 (G) := {x ∈ A : f (x) ∈ G}
uniformly continuous on A and (xn ) is a Cauchy (preimage, not inverse!).
sequence in A, then (f (xn )) is a Cauchy sequence.
Theorem 5.4.8 (Continuous Extension Corollary 11.3.3 A function f : R 7→ R is continuous
Theorem) A function f is uniformly continuous on iff f −1 (G) is open in R, whenever G is open. Or, the
the interval (a, b) iff lim f (x) and lim f (x) can be preimage of an open set is always open.
x7→a+ x7→b−
defined at the endpoints a and b such that the Theorem 5.6.7 If m ∈ Z, n ∈ N and x > 0, then Remark. In general, for an open set G, the direct
1 m 1
Downloaded
extended function is continuous by b].
on [a, Anthony Rrewn ([email protected])
(x n ) = (xm ) n image f (G) is not necessarily open.
 lOMoARcPSD|48850158

Metric Spaces (iii) Exponential f (x) = ax is monotonely State discontinuity criterion.

increasing/decreasing for a > 1 and 0 < a < 1 State combinations/composition of functions.
Definition 11.4.1 A metric on a set S is a function
respectively. State max-min theorem.
d : S × S 7→ R that satisfies the following properties:
sin properties State location of roots and intermediate value
(a) d(x, y) ≥ 0, ∀x, y ∈ S (positivity); (i) sin x ≤ x theorem.
Theorem A subset E of R is connected iff E is an
(b) d(x, y) = 0 ⇐⇒ x = y (definiteness); (ii) sin x − sin y = sin x−y x−y
2 cos 2
Define preservation of closed intervals theorem.
interval. How to prove uniform continuity (4)?
(c) d(x, y) = d(y, x), ∀x, y ∈ S (symmetry); Theorem (Preservation of connectedness) Let
(d) d(x, y) ≤ d(x, z) + d(z, y), ∀x, y, z ∈ S (triangle
List of Questions How to prove non-uniform continuity (2)?
f : S 7→ R be a continuous function. If E is a Define theorems of uniform continuity (2x)
inequality); The Real Numbers
connected subset of S, then f (E) is connected. Define one-sided limits for monotone function exist
Define Bernoulli’s Inequality.
Definition 11.2.1 An open cover of a subset A of a Corollary (Generalised form of preservation of theorem and it’s usefulness.
closed intervals). Let f : S 7→ R be a continuous Define Triangle Inequality and it’s two corrollaries.
metric space S is a collection of G := {Gλ : λ ∈ Λ} of Define jump and it’s usefulness.
function. If K is a connected compact subset of S, The Completeness Property of R
open subsets of S whose union contains A; that is, Define discontinuous points of monotone functions
then f (K) = [inf f (K), sup f (K)], a closed bounded Define −neighborhood of a.
Define sup and inf and sup/inf property. theorem.
interval. Define continuous inverse theorem.
A ⊆ ∪λ∈Λ Gλ Define Archimedean Property and corollary.
Define Density theorem (both rational and State usefulness of rational power functions.
Calculus A glimpse into topology
L’Hospital’s Rule. Suppose that we have one of the irrational).
If G 0 is a subcollection of sets from G, such that the Define neighbourhood, open set, closed set.
following cases, Describe nested interval property.
union of the sets in G 0 also contains A, then G 0 is Define open set properties, closed set properties.
How to prove convergence and calculate limits: five
called a subcover of G. Define characterisation of closed sets and open sets.
f (x) 0 ±∞ ways to identify limits (for cauchy criterion, state
If G 0 consists of finitely many sets, then we call G 0 a lim = OR negation too). Define Global Continuity Theorem corollary
finite subcover of G.
x7→α g(x) 0 ∞ Define metric conditions
Define divergent properties.
Define open cover
Definition 11.2.2 A subset K of a metric space S is where α can be any real number, ±∞. In these cases, State relation between limits of subsequences and
State intuitive understanding of disconnected
said to be compact if every open cover of K has a we have, sequences (2).
State theorems related to compactness (x2) and
finite subcover. State and define five ways to test if infinite series
f (x) f 0 (x) connectedness (x2)
Definition. Let (S, d) be a metric space. A subset A converge/diverge
lim = lim State Generalised form of preservation of closed
of S is called bounded if there exist a positive real x7→α g(x) xtoα g 0 (x) State and define three ways to test for absolute
intervals
number M and a point x ∈ S such that A ⊆ VM (x) convergence.
Partial Fraction Decomposition. (1) Factor the Calculus
(i.e. d(a, x) < M, ∀a ∈ A). State relation between limit of modulus and modulus
What to do when you see n22n+1 ?
bottom. x25x−4 5x−4
= (x−2)(x+1) . of limit (n+1)2
Theorem 11.2.5. Let (S, d) be a metric space. A −x−2
If x < y and 0 < a < 1, what’s the inequality relating
(2) Write one partial fraction for each of those Limits
subset K of S is compact iff it is closed and bounded. ax and ay ?
5x−4 A1 A2 Define cluster point.
Theorem 11.4.13 (Preservation of factors. (x−2)(x+1) = x−2 + x+1 .
State ways to prove limits of functions (3, excluding
Compactness) If (S, d) is a compact metric space (3) Multiply through by the bottom so we no longer one-sided limit) Examples to relook
and if the function f : S 7→ R is continuous, then have fractions. 5x − 4 = A1 (x + 1) + A2 (x − 2). State lower bound theorem. 2.1.13 (1, 2), 2.2.6(b), 2.4.1(b), 3.1.6(d), 3.2.8 (e),
f (S) is compact in R. (4) Now find constants A1 and A2 by substituting Continuous Functions 3.4.3(a), 3.5.6(c), 3.7.6(c), T3Q2 (FD), T3Q5,
Definition (disconnected) A subset U of S is the roots (i.e. x = −1, 2). State ways to show that a function is continuous at a T5Q2(a), T6Q3, 18/19 (Q2), 13/14(Q1(a), Q5(ii))
called disconnected if U has an open cover {A, b} Inequalities point (4)
1
such that A ∩ B ∩ U = ∅, A ∩ U 6= ∅, B ∩ U 6= ∅. (i) Multiplicative Inverse If a < b, then a < 1 Define continuous functions and sequential criterion
√ b √
Otherwise, U is said to be connected. (ii) Square Root Property If a ≤ b, then a ≤ b for continuity. State usefulness of relation to limit. Prepared by Larry, AY2020/2021 Semester 1

Downloaded by Anthony Rrewn ([email protected])