0.1.

ORDER 1

0.1 Order 0.6 Closed

Definition (Order). Let x, y ∈ S. Then x < y, x = y, or Definition (Closure). Closure of E, E, is E
S
E′
x > y.
Definition (Ordered Set). Set with an Order Theorem. The closure of E is closed, E = closure of E iff
E is closed, if E ⊂ F and F is closed then the closure of E
Definition (Bounded Above). Subset E ⊆ S of an or- ⊂F
dered set is bounded above iff ∃B ∈ S such x ≤ B ∀x ∈ E.
Definition (Least upper bound property). For all Theorem. If E is a set of real numbers and supremum of
nonempty bounded above subsets E ⊆ S, sup E ∈ S exists E is real, then supremum of E is in E iff E is closed
Theorem (sup of all lower bounds = ∞). For an or- Theorem. E is open iff E c is closed
dered set S with l.u.b, and let B be the subset that is bounded
below, with L as the set of all lower bounds of B. Then the Theorem. Intersection and unions of closed sets are closed
α = supremum of L exists and α ∈ S with ∞B = α

0.2 Triangle Inequality 0.7 Open Sets

Theorem (Triangle Inequality). |X + Y | ≤ |X| + |Y | Definition (Interior Point). exists a neighborhood

Pn around p such that Nr (p) is in E
Theorem (Schwarz Inequality). | j=1 zj wj |2 ≤
Pn 2
Pn 2
j=1 |zj | ∗ j=1 |wj | Definition (Open Relative). E ⊂ X ⊂ Y is open relative
to X if for all x in E there exists r > 0 such Nr (p) ∩ X ⊂ E

0.3 R and C Theorem. E is open relative to X iff there exists open set
G in Y such G ∩ Y = E
Definition (Field). Closed under +, *; commutative, as-
sociative, identity Theorem. unions and finite intersections of Open sets are
Definition (Complex conjugate x). x = a + bi → x = open
a − bi
Theorem (Existence TH). There exists an ordered field 0.8 Perfect Sets
R such Q is a subfield that has l.u.b
Theorem (Archimedian Property). For all x, y in R, Definition (Perfect). E is closed and all points of E are
with x > 0, there exists natural number n such nx > y limit points of E
Theorem. For all real x > 0 and all natural n there exists
Theorem. Nonempty perfect sets in Rk are uncountable
a unique positive real y such y n = x
Theorem (Cantor Set). Start with [0, 1] and take out the
0.4 Countability middle 1/3 repeatedly

Theorem. Countable union of countable sets are count-

able, and every subset of countable set is countable 0.9 Compact Sets
Theorem. Set of n-tuples of elements in a countable set is
Definition (Compact). All open covers have a finite sub-
countable
cover

0.5 Metric Spaces Theorem. Compact sets are closed, and closed subsets of
compact sets are compact
Definition (Distance). d(x, x) = 0,
Theorem. Intersection of closed sets and compact sets are
d(p, q) > 0,
compact
d(p, q) = d(q, p),
d(p, q) ≤ d(p, x) + d(x, q) Theorem. K ⊆ Y ⊆ X is compact relative to Y iff K is
Definition (Limit Point). A limit point of E, p, is such compact relative to X
that any neighborhood around p contains inftyinitely many
points of E. Theorem. K is compact iff for all infinite subset E of K
has a limit point in K
Definition (Isolated Point). Any point of E that isn’t a
limit point of E Theorem. If every finite intersection of a set of compact
Definition (Dense). For all points in X are in E or limit sets are not empty, then the infinite intersection of the set
points of E is not empty
2

Theorem. For a series of decreasing nested compact sets, 0.13 Continuous Functions
the intersection is not empty.
Definition (Function Limit). limx→p f (x) = q if for all
Theorem. If pn is in a compact space X then some subse- ϵ > 0 there exists δ > 0 such that for all d(x, p)< δ →
quence converges in X d(f(x), q) → ϵ
OR
Theorem. Let kn be a compact set with k1 ⊇ k2 ⊇ ....
limx→p f (x) = q iff limx→p+ f (x) = limx→p− f (x) = q
If diam kn → 0 then the intersection of all these sets is a
singleton Theorem. limx→p f (x) = q iff limn→∞ f (tn ) = q for all
sequence pn such pn → p
Theorem. All compact sets are complete Definition (Continuous). f is continuous at p if
limx→p f (x) = f (p)

0.10 Compactness and Continuity Theorem. f : X → Y is continuous on X iff for all open
sets U ⊂ Y , f −1 (U ) is open in X
Theorem. f : X → Y is continuous implies f is uniformly Theorem. If f is monotonic then for all x in (a, b):
continuous supa<t<x f (t) = f (x− ) ≤ f (x) ≤ f (x+ ) = infx<t<b f (t)

Theorem. In a compact space, if f is bijective then f being

continuous implies its inverse is also continuous 0.14 Discontinuity
Theorem. f is continuous, then f(K) for any compact sub- Definition (First Kind). if f(m+) and f(m-) exists but
sets of X is compact they don’t equal f(m), or if f(m+) ̸= f(m-)
Theorem. f (a, b) → R is monotonic then f is discontinu-
Theorem. If X is compact, and f : X → R is continu-
ous at countably many points
ous, then there exists p,q in X such f(p) = supremum of the
image of f, and f(q) = infinum of the image of f
0.15 Limits at ∞
0.11 Connected Sets Definition (Neighborhoods at ∞). (c, ∞) or (−∞, c)
Definition (f(t) → A). f(t) → A as t → x where A,x
Definition (Separated Set). A ∩ B = ∅ and B ∩ A = ∅ ∈ (R ∪ ∞) if for every neighborhood U of A there is a neigh-
borhood V of x such that V ∩ E is not empty and f(t) ∈ U
Theorem. E in R is connected iff for all x, z in E, if for all t ∈ V ∩ E, t ̸= x
x < y < z then y is in E

Theorem. If f is continuous and E is connected, then f(E) 0.16 Sequence

is connected
Definition (Convergent Sequences). For all ϵ > 0
Theorem (Intermediate Value TH). If f (x) < f (y) there exists a natural number N such for all n ≥ N then
then for all c in [f(x), f(y)] there exists x in [x,y] such f(x)=c d(pn , p) < ϵ
Theorem. for pn ⊆ X:
k every neighborhood of p contains all but finitely
0.12 R many points of pn
limits are unique
Theorem. For a series of decreasing nested K-cells, the
if pn → p then pn is bounded
intersection is not empty.
limits are limit points
Theorem (Heine Borel). Compact iff closed and bounded Theorem. Jf a sequence is monotonic, then it converges
in Rk iff it is bounded
Theorem. Every bounded sequence in R has a convergent
Theorem (Weistrass). Every bounded inftyinite subset of
subsequence.
Rk has a limit point

Theorem (Cauchy Criterion). Every cauchy sequence 0.17 Cauchy Sequences

in Rk is convergent
Definition (Cauchy Sequences). For all ϵ > 0 there ex-
Theorem. If functions f and g are continuous, the f+g, f*g
ists a natural number N such that for all n, m ≥ N, then
are continuous.
d(pn , pm ) < ϵ
Theorem. for xn where xn = a1n , a2n , ... converges to x iff Definition (Diameter of E). supremum of S, where S is
ain → B the set of all distances between points in E.
0.18. SUBSEQUENCES 3

0.18 Subsequences 0.23 Power Series

Theorem. pn → p iff any subsequence pni → p Definition (Power Series).
P
cn z n
∗
Theorem. pn → p iff P = P∗ = p p
Theorem (Power Series Test). α = lim sup n
|cn | and
Theorem. Let E be the set of subsequential limits. Then R = α1
E is closed, both limsup and liminf of the sequence is in E. |Z| < R : Converges
|Z| > R : Diverges
Definition (Alt Def of Lim Sup). S ∗ = sup(E) or
αk = sup(sn |n ≥ k); S ∗ = limk→∞ αk
Theorem. if radius ofPconvergence is 1 and c0 ≥ c1 ≥ ...
Theorem. if sn ≤ tn → s∗ ≤ t∗ and s∗ ≤ t∗ with lim cn = 0, then cn z n converges at all of |z| = 1
except z = 1
Theorem (Baby Squeeze Theorems). if sn ≤ tn for all
n ≥ N , then if tn converges, so does sn
Theorem. if x > s∗ , then there exists natural N such for 0.24 Summation by Parts
all n > N, sn ≤ x
Pn
Definition (Summation by parts). An = k=0 ak ; for
0.19 Special Sequences Pq Pq−1
0≤ p ≤ q: then k=p ak bk = k=p An (bk − bk+1 ) + Aq bq +
Ap−1 bp
Theorem. for p > 0: n1p → 0 √
√
for p > 0: n p → 1 n n → 1 Theorem. If the partial sums of An form a bounded se-
nα
for p > 0, α ∈ R: (1+p)n → 0 bn is monotonically decreasing, and limn→∞ bn →
quence, P
q
for |x| < 1: xn → 0 0, then k=p ak bk converges.

0.20 Series
0.25 Cauchy product
P
Definition (Sum Convergence). ak converges iff for P
all
Pn ϵ > 0 there exists natural N such that for all n,m > N, Definition
Pn (Cauchy Product). Cn where Cn =
| k=m ak | < ϵ a b
k=0 k a−k
PN
Definition (Partial Sum). SN := n=1 an Theorem. P if
P
an P
converges
P
P absolutely, then Cn con-
Theorem.
P
an with an ∈ R≥0 converges iff sequence of verges and Cn = ( an )( bn )
partial sums are bounded P P P
P Theorem. If P an = A and bn = B and Cn con-
Theorem. if an converges then an → 0 verges, then Cn = AB
Theorem (SqueezeP Theorems). if P |sn | ≤ cn for all
n ≥ N , then if cn converges, so does sn
0.26 Rearrangements
0.21 Absolute convergence P
Theorem. If an converges but not absolutely, for any
Definition (Abs Convergence).
P
|an | converges finite α < β, we can find some arrangement such that the
liminf is α and the limsup is β
Theorem. Absolute convergence implies convergence
P
Theorem. If an converges absolutely, then all rear-
rangements converge and to the correct sum
0.22 Convergence Tests
p
Theorem (Root Test). α = lim sup n |an |
α < 1 : Converges 0.27 Special Series
α > 1 : Diverges
P 1 1
Theorem (Ratio Test). α = lim sup | aan+1 | Theorem. for 1 ≥ x ≥ 0: np = 1−x P
α < 1 : Converges
n
P k for a1 ≥ a2 ≥ ...: | an | converges iff
α ≥ 1 : Diverges | 2 a2k | converges
for p > 1,: P n1p converges
P
∞
Theorem. lim sup | aan+1
p 1
n
| ≥ lim sup n |an | for p > 1: n=2 n(log(n))p converges
lim inf | aan+1
p P 1 1 n
n
| ≥ lim inf n |an | e= n! = lim(1 + n )
4

0.28 Differentiation Theorem. If f is continuous on [a,b] then f ∈ ℜ(α) on [a,b]

Theorem. If f is monotonic on [a,b] and α is continuous

Theorem. Differentiable on [a,b] implies continuous on
on [a,b] then f ∈ ℜ(α) on [a,b]
[a,b]
Theorem. if f is bounded on [a,b] and f is not continuous
Theorem (Mean Value TH). f,g are continuous and
at finitely many points where α is continuous, the f ∈ ℜ(α)
differentiable on (a,b). Then there exists x in (a,b) such
(f (b) − f (a))g ′ (x) = (g(b) − g(a))f ′ (x) Theorem. if f ∈ ℜ(α) on [a,b] and f is bounded m ≤
f ≤ M , and suppose g is continuous on [m, M], then
Theorem. if f is differentiable on [a,b], then suppose
h(x)g(f (x)) ∈ ℜ(α) on [a,b]
f ’(a)< λ < f ’(b). Then for some x in [a,b], f ’(x) = λ
Definition (Local Max). p is a local max if there exists
a δ > 0 such that for all q in a δ neighborhood implies f(q) 0.30 Properties of an Integral
≤ f(p) R
f ’(p) = 0 Theorem.
R if f R∈ ℜ(α1 ) and f ∈ ℜ(α2 ) , then f d(α1 +
α2 ) = f dα1 + f dα2
Theorem (L’hopitals). if f,g: (a,b) → R are differen-
tiable with g’(x) ̸= 0 for all x in (a,b), and −∞ ≤ a < Theorem. if f is bounded on [a,b], and s is in (a,b), and f
Rb
b ≤ ∞. Then, if both f(x) and g(x) approaches 0 or g(x) is continuous at s, and α(x) = I(x − s), then a f dα = f(s)
′
approaches ∞, as x → a, then fg(x)
(x)
→ limx→a fg′ (x)
(x)
Theorem. if cn ≥ 0 for all n, and
P
cn converges, with
sP
n as the sequence of distinct points inR (a,b) and P α(x) =
Theorem (Taylor TH). if f: (a,b) → R, f ( n − 1) : cn I(x−sn ). If f is cont on [a,b], then f dα = cn f (sn )
(a, b) → R is continuous, let a ≤ α < β ≤ b, and p(t)
Pn−1 (k)
= k=0 f k!(α) (t − α)k then there exists x in (α, β) such Theorem. If α(x) is monotonically increasing, α′ (x) ∈
ℜ(x) on [a,b], and f is boundedR on [a,b].
R Then, f is in-
f (β) = p(β) + f (n) (x)
− α)n tegrable iff fα′ is integrable and f dα = f α′ dx
n! (β

Theorem (Change of Variables). Suppose g:[A,B] →

0.29 Integration [a,b] is strictly increasing and continuous. Suppose α is
monotonically increasing on [a,b], and f ∈ ℜ(α). Let B(y)
Definition (Partition). P = x0 , x1 , ..., xn s.t. a = x0 ≤ = α(g(y)) and C(y) = f(g(y)) for y∈ [A, B]. Then, C ∈ (B)
Rb Rb
x1 ≤ ... ≤ xn = b and a CdB = a f dα

Definition (Upper and Lower Sums). ∆αi = α(xi ) −

α(xi−1 ) 0.31 Integration and Differentia-
Mi := supf (x),
P mi := inf f (x)
U (p, f, α) = P Mi ∆αi
tion
L(p, f, α) = mi ∆αi
Theorem. Let f ∈ (x) on [a,b]. For x in [a,b], let F(x) =
Upper and lower riemann integrals are the sup/inf of the Rx
a
f (t)dt. Then, F is cont on [a,b], and if f is cont at p,
upper and lower sums over all partitions of [a,b]
then F is diff at p with F’(p) = f(p)
Definition (Integrable w. respect to α). if lower rie-
Theorem (Fund Theorem of Calc). Suppose f ∈ (x) on
mann integral = upper riemann integral
[a,b]. If there exists a diff function F on [a,b] such F’ = f
Rb
Theorem. Suppose P ⊂ P ∗ . Then, L(P, f, α) ≤ then a f (x)dx = F (b) − F (a)
L(P ∗ , f, α) ≤ U (P ∗ , f, α) ≤ U (P, f, α)
Theorem (Integration by Parts). If F,G diff with F’, G’
∈ (x), then F (x)G′ (x)dx + F ′ (x)G(x)dx = F (b)G(b) −
R R
Theorem. If f ∈ ℜ(α) on [a,b] iff for all ϵ > 0 there exists
a partition such U (P, f, α) − L(, f, α) < ϵ F (a)G(a)