IB Math HL (AA/AI) · ClipSAT

⏱ Exam Timer: 00:00
IB · International Baccalaureate

IB Mathematics HL

Courses · Analysis & Approaches (AA) · Applications & Interpretation (AI) Exam · Paper 1 (no calc) · Paper 2 (GDC) · Paper 3 (extended investigation) Topics · 5 core topics + HL extensions
i

Overview

About IB Mathematics HL

Analysis and Approaches HL (AA HL) is the most rigorous IB math course — equivalent to a first-year university pure mathematics course. It extends AA SL with complex numbers, proof by induction, vectors in 3D, advanced integration (by parts, substitution, partial fractions), differential equations, and series convergence.

Applications and Interpretation HL (AI HL) extends AI SL with matrices, complex numbers, Poisson distribution, graph theory (networks), and deeper optimization and differential equations.

IB HL Syllabus Map (AA & AI)

TopicAA HL specificsAI HL specifics~Marks
1 — AlgebraProof by induction, complex numbersMatrices, eigenvalues20
2 — FunctionsModulus, complex roots, partial fractionsAdditional regression20
3 — Geometry & Trig3D vectors, cross product, planesGraph theory, algorithms25
4 — StatisticsUnbiased estimators, CIs, hypothesis testsPoisson, Markov chains30
5 — CalculusBy parts, DEs, Maclaurin seriesCoupled DEs, numerics40

Exam: Paper 1 (no GDC, 120 min) + Paper 2 (GDC, 120 min) + Paper 3 (extended, 60 min). Internal Assessment 20%.

IB HL — Every subtopic, linked to its lesson

TopicSubtopicLesson
1 — Number & AlgebraCore: sequences, exponents, logs, binomial theorem, proof, financial mathsSL Ch 1 — Number & Algebra
1 — Number & AlgebraProof by mathematical induction (AA HL)Ch 1 — Algebra, Proof & Complex Numbers
1 — Number & AlgebraComplex numbers & the Argand plane (AA HL)Ch 1 — Algebra, Proof & Complex Numbers
1 — Number & AlgebraPolar form & de Moivre's theorem (AA HL)Ch 1 — Algebra, Proof & Complex Numbers
1 — Number & AlgebraPartial fractions (AA HL)Ch 1 — Algebra, Proof & Complex Numbers
1 — Number & AlgebraSystems of linear equations (AA HL)Ch 1 — Algebra, Proof & Complex Numbers
1 — Number & AlgebraConvergence of series (AA HL)Ch AA — Additional Topics
1 — Number & AlgebraMatrices (AI HL)Ch AI — Additional Topics
2 — FunctionsCore: domain/range, composite, inverse, transformationsSL Ch 2 — Functions
2 — FunctionsModulus function & solving |f(x)| = k (AA HL)Ch 2 — Functions (HL Extensions)
2 — FunctionsFactor & remainder theorem (AA HL)Ch 2 — Functions (HL Extensions)
2 — FunctionsGraphs of inverse trigonometric functions (AA HL)Ch AA — Additional Topics
3 — Geometry & TrigonometryCore: circle, triangle trigonometry, 2D vectorsSL Ch 3 — Geometry & Trig
3 — Geometry & TrigonometryCompound angle identities & reciprocal trig functions (AA HL)Ch 3 — Vectors & 3D Geometry
3 — Geometry & Trigonometry3D vectors & the cross product (AA HL)Ch 3 — Vectors & 3D Geometry
3 — Geometry & TrigonometryLines and planes in 3D (AA HL)Ch 3 — Vectors & 3D Geometry
4 — Statistics & ProbabilityCore: descriptive statistics, probability, Binomial & NormalSL Ch 4 — Statistics & Probability
4 — Statistics & ProbabilityUnbiased estimators (AI HL)Ch 4 — Statistics & Probability (HL)
4 — Statistics & ProbabilityConfidence intervals for the mean (AI HL)Ch 4 — Statistics & Probability (HL)
4 — Statistics & ProbabilityPoisson distribution (AI HL)Ch 4 — Statistics & Probability (HL)
4 — Statistics & ProbabilityMarkov chains (AI HL)Ch AI — Additional Topics
5 — CalculusCore: differentiation, integration, kinematicsSL Ch 5 — Calculus
5 — CalculusFurther integration techniques (AA HL)Ch 5 — Calculus (HL)
5 — CalculusDifferential equations (AA HL)Ch 5 — Calculus (HL)
5 — CalculusMaclaurin series (AA HL)Ch 5 — Calculus (HL)
5 — CalculusGraph theory & Kruskal's algorithm (AI HL)Ch AI — Additional Topics

Every official HL subtopic above links straight to the lesson that covers it — "Core" rows point to the shared SL lesson (same content at HL), everything else to this track's own HL-extension chapters. Click any lesson name to jump there.

Topic 1 HL
Proof by induction, complex numbers, partial fractions, sum of series, system of linear equations.
Topic 2 HL
Rational functions, polynomial long division, modulus function, odd/even, complex roots.
Topic 3 HL
3D vectors, lines/planes in 3D, cross product, intersections, angles between planes.
Topic 4 HL
Poisson distribution (AI), unbiased estimators, confidence intervals, hypothesis testing.
Topic 5 HL
Integration by parts, substitution, partial fractions, differential equations (separable, homogeneous), Maclaurin series.

Exam (both courses): Paper 1 (no GDC, 120 min, 110 marks) + Paper 2 (GDC, 120 min, 110 marks) + Paper 3 (60 min, 55 marks — investigative, GDC required).

⚠ IB Mathematics HL Exam Traps & Command Terms
Accuracy Standard: Non-exact numerical answers must be given to exactly 3 significant figures unless specified otherwise. Intermediate rounding errors incur an automatic accuracy mark penalty (A1).

Command Terms: 'Show that' means the answer is already provided; every single intermediate algebraic step must be justified without skipping. 'Hence' means you must use the preceding part's result; alternate methods receive no credit.

Paper 1 vs Paper 2: Paper 1 prohibits any calculator; you must show exact values involving fractions, surds, ee, and π\pi. Paper 2 and 3 require GDC fluency (graphing, matrix solvers, numerical integration, and statistics distributions).
Bar chart: IB Mathematics HL Assessment Weighting (%) with categories: Paper 1 (non-calc) (30), Paper 2 (calc) (30), Paper 3 (investigation) (20), Internal Assessment (20)08152330Paper 1 (non-calc)30Paper 2 (calc)30Paper 3 (investigation)20Internal Assessment20IB Mathematics HL Assessment Weighting (%)
IB Mathematics HL assessment component weighting: Paper 1 (30%), Paper 2 (30%), Paper 3 (20%), and Internal Assessment (20%).
A student receives the following scaled marks on IB AA HL assessment: Paper 1: 85/110, Paper 2: 92/110, Paper 3: 44/55, and IA: 16/20. Calculate the student's overall weighted percentage.
1. Compute the component percentages:
Paper 1: 85110≈77.27%\dfrac{85}{110} \approx 77.27\%
Paper 2: 92110≈83.64%\dfrac{92}{110} \approx 83.64\%
Paper 3: 4455=80.00%\dfrac{44}{55} = 80.00\%
IA: 1620=80.00%\dfrac{16}{20} = 80.00\%

2. Multiply by component weights (30%, 30%, 20%, 20%):
Overall=0.30(77.27)+0.30(83.64)+0.20(80.00)+0.20(80.00) \text{Overall} = 0.30(77.27) + 0.30(83.64) + 0.20(80.00) + 0.20(80.00)
Overall=23.181+25.091+16.000+16.000=80.27%≈80.3%. \text{Overall} = 23.181 + 25.091 + 16.000 + 16.000 = 80.27\% \approx 80.3\%.
📝 Chapter Quiz
1

Algebra, Proof & Complex Numbers

IB Mathematics HL P1, P2 & P3 AA HL AI HL

HL Topic 1 extends SL algebra with proof by induction and complex numbers — the foundation for much of higher mathematics.

● Proof by mathematical induction

Structure: (1) Base case: verify for n=1n=1 (or initial nn). (2) Inductive hypothesis: assume true for n=kn=k. (3) Inductive step: prove true for n=k+1.n=k+1. (4) Conclude.

● Complex numbers

z=a+bi,i2=−1,Re(z)=a,Im(z)=b.z=a+bi,\quad i^2=-1,\quad\text{Re}(z)=a,\quad\text{Im}(z)=b.∣z∣=a2+b2,arg⁡(z)=arctan⁡(b/a) (adjusted for quadrant).|z|=\sqrt{a^2+b^2},\quad\arg(z)=\arctan(b/a)\text{ (adjusted for quadrant)}.zˉ=a−bi,zzˉ=∣z∣2,1z=zˉ∣z∣2.\bar{z}=a-bi,\quad z\bar{z}=|z|^2,\quad\frac{1}{z}=\frac{\bar{z}}{|z|^2}.

⊙ Explorer · Complex numbers on the Argand planeinteractive
|z|5.000
arg(z)53.13°
z̄ (conjugate)3.00 − 4.00i

z = a + bi (indigo dot) and its conjugate z̄ = a − bi (amber, dashed segment) are mirror images across the real axis — exactly what negating the imaginary part does geometrically. |z| is the dot's distance from the origin; arg(z) is the angle the segment from the origin to z makes with the positive real axis, drawn as an arc. Drag a or b negative to move z into another quadrant and watch arg(z) jump to the correctly-adjusted value, not just arctan(b/a) taken literally.

● Polar form and de Moivre's theorem

z=r(cos⁡θ+isin⁡θ)=reiθ(r=∣z∣,  θ=arg⁡z).z=r(\cos\theta+i\sin\theta)=re^{i\theta}\quad(r=|z|,\;\theta=\arg z).zn=rn(cos⁡nθ+isin⁡nθ)(de Moivre’s theorem).z^n=r^n(\cos n\theta+i\sin n\theta)\quad(\text{de Moivre's theorem}).

nn-th roots of unity: solutions to zn=1z^n=1 are e2πik/ne^{2\pi ik/n} for k=0,1,…,n−1.k=0,1,\ldots,n-1.

⊙ Explorer · De Moivre's theorem on the unit circleinteractive
zⁿcis(90.0°)
root spacing (360°/n)120.0°
rⁿ = |zⁿ|1.000

z = r·cis θ (indigo) and zⁿ = rⁿ·cis(nθ) (amber, dashed) — this is de Moivre's theorem in full: raising z to the nth power multiplies the angle by n AND raises the modulus to the nth power. The gray rings are the n-th roots of unity (always on the unit circle, r = 1), k = 0, …, n−1, evenly spaced 360°/n apart starting at angle 0°. Set r = 1 and drag θ to a multiple of that spacing to land zⁿ exactly on z = 1, the k = 0 root. Push r above or below 1 and watch zⁿ leave the unit circle — outward for r > 1, inward for r < 1 — since |zⁿ| = rⁿ grows or shrinks with n, unlike the roots of unity themselves.

● Partial fractions (AA HL)

p(x)(x−a)(x−b)=Ax−a+Bx−b,p(x)(x−a)(x2+b)=Ax−a+Bx+Cx2+b.\frac{p(x)}{(x-a)(x-b)}=\frac{A}{x-a}+\frac{B}{x-b},\quad\frac{p(x)}{(x-a)(x^2+b)}=\frac{A}{x-a}+\frac{Bx+C}{x^2+b}.

● Systems of linear equations (AA HL)

A system of up to three linear equations in three unknowns can be solved by elimination: combine pairs of equations to cancel one variable at a time, reducing to a smaller system, until a single equation in one unknown remains — then substitute back up.

Geometrically, each equation ax+by+cz=dax+by+cz=d is a plane in 3D (Chapter 3 of this track studies these directly). Solving the system asks how three planes intersect:

  • Unique solution — the three planes meet at exactly one point (the generic case).
  • No solution — elimination produces a contradiction like 0=k0=k for k≠0k\neq0; geometrically, no point lies on all three planes at once (e.g. two are parallel, or the three form a triangular "tube" with no common point).
  • Infinitely many solutions — elimination produces an identity like 0=00=0, leaving a free parameter; geometrically, the planes share a common line (or are all the same plane).

Worked Example 1.4 — Solving a 3×3 system by elimination

Problem. Solve: x+y+z=6x+y+z=6, 2x−y+z=32x-y+z=3, x+2y−z=2x+2y-z=2.

Label the equations (1), (2), (3). Add (1) and (3) to eliminate zz: (x+y+z)+(x+2y−z)=6+2⇒2x+3y=8(x+y+z)+(x+2y-z)=6+2\Rightarrow2x+3y=8 — call this (4).

Subtract (2) from (1) to eliminate zz again: (x+y+z)−(2x−y+z)=6−3⇒−x+2y=3⇒x=2y−3(x+y+z)-(2x-y+z)=6-3\Rightarrow-x+2y=3\Rightarrow x=2y-3 — call this (5).

Substitute (5) into (4): 2(2y−3)+3y=8⇒4y−6+3y=8⇒7y=14⇒y=2.2(2y-3)+3y=8\Rightarrow4y-6+3y=8\Rightarrow7y=14\Rightarrow y=2.

Then x=2(2)−3=1x=2(2)-3=1, and from (1): 1+2+z=6⇒z=3.1+2+z=6\Rightarrow z=3.

Check in (2): 2(1)−2+3=3.2(1)-2+3=3. ✓ Check in (3): 1+2(2)−3=2.1+2(2)-3=2. ✓

Solution: (x,y,z)=(1,2,3)(x,y,z)=(1,2,3) — the unique point where all three planes meet.

Worked Example 1.1 — Proof by induction

Problem. Prove by induction: ∑r=1nr2=n(n+1)(2n+1)6.\displaystyle\sum_{r=1}^n r^2=\frac{n(n+1)(2n+1)}{6}.

Base (n=1n=1): LHS=1.=1. RHS=1⋅2⋅36=1.  ✓=\frac{1\cdot2\cdot3}{6}=1.\;\checkmark

Assume ∑r=1kr2=k(k+1)(2k+1)6.\displaystyle\sum_{r=1}^k r^2=\frac{k(k+1)(2k+1)}{6}.

For n=k+1n=k+1: ∑r=1k+1r2=k(k+1)(2k+1)6+(k+1)2=(k+1) ⁣[k(2k+1)6+(k+1)]\displaystyle\sum_{r=1}^{k+1}r^2=\frac{k(k+1)(2k+1)}{6}+(k+1)^2=(k+1)\!\left[\frac{k(2k+1)}{6}+(k+1)\right]

=(k+1)⋅2k2+7k+66=(k+1)(k+2)(2k+3)6.  ✓=(k+1)\cdot\frac{2k^2+7k+6}{6}=\frac{(k+1)(k+2)(2k+3)}{6}.\;\checkmark

Worked Example 1.2 — Complex numbers

Problem. Write z=1+i3z=1+i\sqrt{3} in polar form and find z6.z^6.

∣z∣=1+3=2.arg⁡z=arctan⁡(3/1)=π/3.|z|=\sqrt{1+3}=2.\quad\arg z=\arctan(\sqrt{3}/1)=\pi/3.

z=2eiπ/3.z6=26ei⋅6π/3=64 ei2π=64(cos⁡2π+isin⁡2π)=64.z=2e^{i\pi/3}.\quad z^6=2^6e^{i\cdot6\pi/3}=64\,e^{i2\pi}=64(\cos2\pi+i\sin2\pi)=64.

Worked Example 1.3 — Partial fractions

Problem. Decompose 5x−3(x+1)(x−2)\dfrac{5x-3}{(x+1)(x-2)}.

5x−3(x+1)(x−2)=Ax+1+Bx−2.\dfrac{5x-3}{(x+1)(x-2)}=\dfrac{A}{x+1}+\dfrac{B}{x-2}.

Multiply through: 5x−3=A(x−2)+B(x+1).5x-3=A(x-2)+B(x+1).

x=2:  7=3B⇒B=7/3.x=−1:  −8=−3A⇒A=8/3.x=2:\;7=3B\Rightarrow B=7/3.\quad x=-1:\;-8=-3A\Rightarrow A=8/3.

Answer: 8/3x+1+7/3x−2.\dfrac{8/3}{x+1}+\dfrac{7/3}{x-2}.

Practice Problem

Express z=−1+i3z=-1+i\sqrt{3} in polar form and find z6.z^6.

Show solution

r=2,θ=2π3.z=2ei2π/3.r=2,\quad\theta=\frac{2\pi}{3}.\quad z=2e^{i2\pi/3}.

z6=26ei4π=64(cos⁡4π+isin⁡4π)=64.z^6=2^6e^{i4\pi}=64(\cos4\pi+i\sin4\pi)=64.

Worked Example  ·  Complex number in polar form

Express z=−3+iz = -\sqrt{3}+i in modulus-argument form.

Solution: ∣z∣=3+1=2|z|=\sqrt{3+1}=2. arg⁡(z)=π−arctan⁡ ⁣(13)=π−π6=5π6\arg(z)=\pi-\arctan\!\left(\dfrac{1}{\sqrt{3}}\right)=\pi-\dfrac{\pi}{6}=\dfrac{5\pi}{6}.

z=2(cos⁡5π6+isin⁡5π6)=2ei5π/6z = 2\left(\cos\dfrac{5\pi}{6}+i\sin\dfrac{5\pi}{6}\right) = 2e^{i5\pi/6}

Worked Example  ·  Proof by induction

Prove ∑r=1nr=n(n+1)2\sum_{r=1}^n r = \dfrac{n(n+1)}{2} by induction.

Solution: Base: n=1n=1: LHS=1=1, RHS=22=1=\frac{2}{2}=1. ✓

Step: Assume true for n=kn=k. Then ∑r=1k+1r=k(k+1)2+(k+1)=(k+1)(k+2)2\sum_{r=1}^{k+1}r = \frac{k(k+1)}{2}+(k+1)=\frac{(k+1)(k+2)}{2}. ✓

Practice: Find the four fourth roots of 16i16i in polar form.
16i=16eiπ/216i=16e^{i\pi/2}. Roots: 2ei(π/8+kπ/2)2e^{i(\pi/8+k\pi/2)} for k=0,1,2,3k=0,1,2,3.

Worked Example  ·  A system with infinitely many solutions

Solve: x+z=2x+z=2, y=1y=1, x+y+z=3x+y+z=3.

Solution: The third equation is exactly the sum of the first two ((x+z)+y=2+1=3(x+z)+y=2+1=3) — it carries no new information.

From the first two: z=2−xz=2-x, y=1y=1, with xx free. Infinitely many solutions: (x,y,z)=(t, 1, 2−t)(x,y,z)=(t,\,1,\,2-t) for any t∈Rt\in\mathbb{R} — a whole line, exactly the "shared line" case above.

Triangle Origin (0,0)Re(z) = 1z = 1 + i√3 with sides 1, √3, r = 2 and angles π/3; right angle at Re(z) = 1Origin (0,0)Re(z) = 1z = 1 + i√31√3r = 2π/3
Polar form of complex number z = 1 + i√3 = 2·cis(π/3) plotted in the complex plane with modulus r = 2 and argument θ = π/3.
⚠ Topic 1 Exam Traps: Proof by Induction & Complex Roots
Induction Conclusion Statement: IB mark schemes reserve an explicit final mark (R1) for the full concluding sentence: 'Since true for n=1, and if true for n=k then true for n=k+1, by mathematical induction the statement is true for all n ∈ ℤ⁺.' Omitting this loses the mark.

Principal Argument Interval: In IB AA HL, the argument must satisfy −π<Arg⁡(z)≤π-\pi < \operatorname{Arg}(z) \leq \pi. Angles like 300∘300^\circ or 5π3\dfrac{5\pi}{3} must be rewritten as −π3-\dfrac{\pi}{3}.

Partial Fraction Degree Check: For improper rational fractions where degree of numerator ≥\geq degree of denominator, you must perform polynomial division before setting up partial fractions.
📝 Chapter Quiz
2

Functions (HL Extensions)

IB Mathematics HL P1, P2 & P3 AA HL AI HL

HL functions extend SL content with the modulus function, rational functions with oblique asymptotes, and complex roots of polynomials.

● Modulus (absolute value) function

∣f(x)∣: reflect negative part above x-axis.|f(x)|:\text{ reflect negative part above }x\text{-axis}.f(∣x∣): reflect positive part across y-axis.f(|x|):\text{ reflect positive part across }y\text{-axis}.

Equations ∣f(x)∣=k|f(x)|=k give two cases: f(x)=kf(x)=k and f(x)=−k.f(x)=-k. Inequalities ∣f(x)∣<k|f(x)|<k give −k<f(x)<k.-k<f(x)<k.

⊙ Explorer · y = f(x) vs y = |f(x)|, and solving |f(x)| = kinteractive
real solutions in this window1

f(x) = 0.5x³ − 2x (faint grey when |f(x)| is shown, so you can see exactly which parts got reflected). Tick the checkbox to fold the curve's negative dips up above the x-axis, matching |f(x)|'s definition above. The amber line y = k marks where you'd read off solutions to f(x) = k (or |f(x)| = k) — count its crossings with the displayed curve; that count is exactly what the readout reports, live.

● Factor/remainder theorem (polynomials)

Complex conjugate root theorem: if p(x)p(x) has real coefficients and a+bia+bi is a root, then a−bia-bi is also a root.

Relation between roots and coefficients for az3+bz2+cz+d=0az^3+bz^2+cz+d=0 with roots α,β,γ\alpha,\beta,\gamma:

α+β+γ=−b/a,αβ+βγ+γα=c/a,αβγ=−d/a.\alpha+\beta+\gamma=-b/a,\quad\alpha\beta+\beta\gamma+\gamma\alpha=c/a,\quad\alpha\beta\gamma=-d/a.

Worked Example 2.1

Problem. Sketch y=∣2x−4∣y=|2x-4| and solve ∣2x−4∣=x+1.|2x-4|=x+1.

V-shape with vertex at (2,0).(2,0).

Case 1: 2x−4=x+1⇒x=5.2x-4=x+1\Rightarrow x=5. Check: ∣6∣=6=6.  ✓|6|=6=6.\;\checkmark

Case 2: 2x−4=−(x+1)⇒3x=3⇒x=1.2x-4=-(x+1)\Rightarrow3x=3\Rightarrow x=1. Check: ∣−2∣=2=2.  ✓|{-2}|=2=2.\;\checkmark

Worked Example 2.2 — Complex roots

Problem. p(x)=x3−3x2+7x−5p(x)=x^3-3x^2+7x-5 has root x=1.x=1. Find all roots.

Factor out (x−1)(x-1): p(x)=(x−1)(x2−2x+5).p(x)=(x-1)(x^2-2x+5).

Quadratic: x=2±4−202=1±2i.x=\dfrac{2\pm\sqrt{4-20}}{2}=1\pm2i.

Roots: x=1,  1+2i,  1−2i.x=1,\;1+2i,\;1-2i.

Modulus function
∣f(x)∣ |f(x)| reflects negative output above x x -axis
Odd/even symmetry
Odd: f(−x)=−f(x) f(-x)=-f(x) ; Even: f(−x)=f(x) f(-x)=f(x)
Practice Problem

p(x)=x4−2x3+5x2−8x+4p(x)=x^4-2x^3+5x^2-8x+4 has x=1x=1 as a double root. Find all roots.

Show solution

(x−1)2(x-1)^2 divides p(x)p(x). Divide: p(x)=(x−1)2(x2+4).p(x)=(x-1)^2(x^2+4).

x2+4=0⇒x=±2i.x^2+4=0\Rightarrow x=\pm2i. Roots: 11 (double), ±2i.\pm2i.

Worked Example  ·  Rational function sketch

Sketch f(x)=2x2−x−6x2−4f(x)=\dfrac{2x^2-x-6}{x^2-4}, stating all asymptotes and intercepts.

Solution: Factor: f(x)=(2x+3)(x−2)(x+2)(x−2)=2x+3x+2f(x)=\dfrac{(2x+3)(x-2)}{(x+2)(x-2)}=\dfrac{2x+3}{x+2} for x≠2x\neq2.

Vertical asymptote: x=−2x=-2. Horizontal: y→2y\to2. Hole at x=2x=2, y=74y=\frac{7}{4}. xx-int: x=−32x=-\frac{3}{2}. yy-int: y=32y=\frac{3}{2}.

Worked Example  ·  Solving modulus equation

Solve ∣2x−3∣=x+1|2x-3|=x+1.

Solution: Case 1: 2x−3=x+1  ⟹  x=42x-3=x+1\implies x=4. Check: ∣5∣=5|5|=5. ✓

Case 2: 2x−3=−(x+1)  ⟹  3x=2  ⟹  x=232x-3=-(x+1)\implies 3x=2\implies x=\frac{2}{3}. Check: ∣−53∣=53=23+1|{-\frac{5}{3}}|=\frac{5}{3}=\frac{2}{3}+1. ✓

Coordinate plane with graph of f(x) = |2x - 3|; marked points Vertex (1.5, 0) (1.5, 0), y-int (0, 3) (0, 3)-11234246xyf(x) = |2x − 3|Vertex (1.5, 0)y-int (0, 3)
Graph of modulus function f(x) = |2x - 3| showing vertex cusp at x = 1.5 and reflection of negative values across the x-axis.
⚠ Topic 2 Exam Traps: Modulus & Inverse Functions
Squaring Modulus Equations: While squaring both sides of ∣ax+b∣=∣cx+d∣|ax+b| = |cx+d| is convenient, squaring when one side is variable without modulus (e.g. ∣2x−3∣=x−1|2x-3| = x-1) can introduce extraneous roots. Always check candidates in the un-squared equation.

Self-Inverse Functions: For rational functions of the form f(x)=ax+bcx−af(x) = \dfrac{ax+b}{cx-a}, f(x)=f−1(x)f(x) = f^{-1}(x). Recognize this symmetry immediately to save algebraic time.

Domain Restrictions on Inverses: For quadratic functions, an inverse function only exists if the domain is restricted to one side of the vertex (e.g. x≥hx \geq h). Stating f−1f^{-1} without stating its restricted domain x≥kx \geq k loses marks.
Practice: Find the range of f(x)=xx2+1f(x)=\dfrac{x}{x^2+1}.
Let y=xx2+1y=\frac{x}{x^2+1}. Then yx2−x+y=0yx^2-x+y=0. For real xx: Δ=1−4y2≥0  ⟹  y∈[−12,12]\Delta=1-4y^2\geq0\implies y\in\left[-\frac{1}{2},\frac{1}{2}\right].
📝 Chapter Quiz
3

Vectors & 3D Geometry

IB Mathematics HL P1, P2 & P3 AA HL AI HL

HL extends vectors to three dimensions, covering lines and planes in space, the cross product, and intersection problems — foundational for physics and engineering.

● Compound angle identities & reciprocal trigonometric functions (AA HL)

The compound angle formulae extend SL's double-angle identities (Chapter 3 of the SL track) to a sum or difference of any two angles: sin⁡(A±B)=sin⁡Acos⁡B±cos⁡Asin⁡B,cos⁡(A±B)=cos⁡Acos⁡B∓sin⁡Asin⁡B, \sin(A\pm B)=\sin A\cos B\pm\cos A\sin B,\qquad\cos(A\pm B)=\cos A\cos B\mp\sin A\sin B, tan⁡(A±B)=tan⁡A±tan⁡B1∓tan⁡Atan⁡B. \tan(A\pm B)=\frac{\tan A\pm\tan B}{1\mp\tan A\tan B}. Setting A=B=θA=B=\theta in the first two recovers exactly the SL double-angle identities sin⁡2θ=2sin⁡θcos⁡θ\sin2\theta=2\sin\theta\cos\theta and cos⁡2θ=cos⁡2θ−sin⁡2θ\cos2\theta=\cos^2\theta-\sin^2\theta — those are this chapter's special case, not a separate fact.

The reciprocal trigonometric functions are sec⁡θ=1cos⁡θ,csc⁡θ=1sin⁡θ,cot⁡θ=1tan⁡θ=cos⁡θsin⁡θ. \sec\theta=\frac{1}{\cos\theta},\qquad\csc\theta=\frac{1}{\sin\theta},\qquad\cot\theta=\frac{1}{\tan\theta}=\frac{\cos\theta}{\sin\theta}. Dividing the Pythagorean identity sin⁡2θ+cos⁡2θ=1\sin^2\theta+\cos^2\theta=1 through by cos⁡2θ\cos^2\theta (then by sin⁡2θ\sin^2\theta) gives two more standard identities: 1+tan⁡2θ=sec⁡2θ,1+cot⁡2θ=csc⁡2θ. 1+\tan^2\theta=\sec^2\theta,\qquad1+\cot^2\theta=\csc^2\theta.

Worked Example  ·  An exact value via compound angles

Find the exact value of sin⁡75°\sin75° using a compound angle formula.

Solution: Write 75°=45°+30°75°=45°+30°, two angles with known exact sine/cosine values: sin⁡75°=sin⁡(45°+30°)=sin⁡45°cos⁡30°+cos⁡45°sin⁡30°. \sin75°=\sin(45°+30°)=\sin45°\cos30°+\cos45°\sin30°.

=22⋅32+22⋅12=64+24=6+24. =\frac{\sqrt2}{2}\cdot\frac{\sqrt3}{2}+\frac{\sqrt2}{2}\cdot\frac12=\frac{\sqrt6}{4}+\frac{\sqrt2}{4}=\frac{\sqrt6+\sqrt2}{4}.

Worked Example  ·  Proving a reciprocal identity

Prove 1+cot⁡2θ=csc⁡2θ1+\cot^2\theta=\csc^2\theta.

Solution: Start from sin⁡2θ+cos⁡2θ=1\sin^2\theta+\cos^2\theta=1 and divide every term by sin⁡2θ\sin^2\theta (valid whenever sin⁡θ≠0\sin\theta\neq0):

sin⁡2θsin⁡2θ+cos⁡2θsin⁡2θ=1sin⁡2θ  ⟹  1+cot⁡2θ=csc⁡2θ. \frac{\sin^2\theta}{\sin^2\theta}+\frac{\cos^2\theta}{\sin^2\theta}=\frac{1}{\sin^2\theta}\;\Longrightarrow\;1+\cot^2\theta=\csc^2\theta. ✓

● 3D vectors

v=(v1v2v3),∣v∣=v12+v22+v32,u⋅v=u1v1+u2v2+u3v3.\mathbf{v}=\begin{pmatrix}v_1\\v_2\\v_3\end{pmatrix},\quad|\mathbf{v}|=\sqrt{v_1^2+v_2^2+v_3^2},\quad\mathbf{u}\cdot\mathbf{v}=u_1v_1+u_2v_2+u_3v_3.

● Cross product

u×v=∣ijku1u2u3v1v2v3∣=(u2v3−u3v2u3v1−u1v3u1v2−u2v1).\mathbf{u}\times\mathbf{v}=\begin{vmatrix}\mathbf{i}&\mathbf{j}&\mathbf{k}\\u_1&u_2&u_3\\v_1&v_2&v_3\end{vmatrix}=\begin{pmatrix}u_2v_3-u_3v_2\\u_3v_1-u_1v_3\\u_1v_2-u_2v_1\end{pmatrix}.

u×v\mathbf{u}\times\mathbf{v} is perpendicular to both u\mathbf{u} and v.∣u×v∣=∣u∣∣v∣sin⁡θ=\mathbf{v}.\quad|\mathbf{u}\times\mathbf{v}|=|\mathbf{u}||\mathbf{v}|\sin\theta= area of parallelogram.

● Lines and planes in 3D

Line: r=a+td.\mathbf{r}=\mathbf{a}+t\mathbf{d}. Parametric: x=a1+td1,  y=a2+td2,  z=a3+td3.x=a_1+td_1,\;y=a_2+td_2,\;z=a_3+td_3.

Plane: r⋅n^=d,\mathbf{r}\cdot\hat{\mathbf{n}}=d, or ax+by+cz=d,ax+by+cz=d, where n=⟨a,b,c⟩\mathbf{n}=\langle a,b,c\rangle is the normal vector.

Angle between planes: cos⁡θ=∣n1⋅n2∣∣n1∣∣n2∣.\cos\theta=\dfrac{|\mathbf{n}_1\cdot\mathbf{n}_2|}{|\mathbf{n}_1||\mathbf{n}_2|}.

Distance from point PP to plane ax+by+cz=dax+by+cz=d: ∣aPx+bPy+cPz−d∣a2+b2+c2.\dfrac{|aP_x+bP_y+cP_z-d|}{\sqrt{a^2+b^2+c^2}}.

Worked Example 3.1 — Cross product

Problem. Find u×v\mathbf{u}\times\mathbf{v} for u=⟨1,2,3⟩,  v=⟨4,−1,2⟩.\mathbf{u}=\langle1,2,3\rangle,\;\mathbf{v}=\langle4,-1,2\rangle.

u×v=(2(2)−3(−1)3(4)−1(2)1(−1)−2(4))=(4+312−2−1−8)=(710−9).\mathbf{u}\times\mathbf{v}=\begin{pmatrix}2(2)-3(-1)\\3(4)-1(2)\\1(-1)-2(4)\end{pmatrix}=\begin{pmatrix}4+3\\12-2\\-1-8\end{pmatrix}=\begin{pmatrix}7\\10\\-9\end{pmatrix}.

⊙ Explorer · The cross product, in 3Dinteractive
u × v⟨7, 10, -9⟩
|u × v|15.166

u=⟨1,2,3⟩\mathbf u=\langle1,2,3\rangle fixed (blue), v=⟨v1,−1,v3⟩\mathbf v=\langle v_1,-1,v_3\rangle (amber, both components adjustable). u×v\mathbf u\times\mathbf v (green) always stays perpendicular to both — drag either slider and confirm it never tilts away from that right angle.

Worked Example 3.2 — Intersection of planes

Problem. Find the line of intersection of planes π1:x+2y−z=3\pi_1:x+2y-z=3 and π2:2x−y+3z=1.\pi_2:2x-y+3z=1.

Direction of line: n1×n2=⟨1,2,−1⟩×⟨2,−1,3⟩=⟨6−1,−3−2,−1−4⟩=⟨5,−5,−5⟩∝⟨1,−1,−1⟩.\mathbf{n}_1\times\mathbf{n}_2=\langle1,2,-1\rangle\times\langle2,-1,3\rangle=\langle6-1,-3-2,-1-4\rangle=\langle5,-5,-5\rangle\propto\langle1,-1,-1\rangle.

Find a point: set z=0z=0: x+2y=3,  2x−y=1x+2y=3,\;2x-y=1. Solve: x=1,y=1.x=1,y=1. Point (1,1,0).(1,1,0).

Line: r=(110)+t(1−1−1).\mathbf{r}=\begin{pmatrix}1\\1\\0\end{pmatrix}+t\begin{pmatrix}1\\-1\\-1\end{pmatrix}.

Line in 3D
r=a+td \mathbf{r}=\mathbf{a}+t\mathbf{d}
Plane equation
r⋅n=a⋅n \mathbf{r}\cdot\mathbf{n}=\mathbf{a}\cdot\mathbf{n} or ax+by+cz=d ax+by+cz=d
Practice Problem

Find the equation of the plane through A(1,2,3)A(1,2,3), B(4,0,1)B(4,0,1), C(2,3,0)C(2,3,0).

Show solution

AB→=⟨3,−2,−2⟩,  AC→=⟨1,1,−3⟩.\overrightarrow{AB}=\langle3,-2,-2\rangle,\;\overrightarrow{AC}=\langle1,1,-3\rangle.

n=AB→×AC→=⟨8,7,5⟩.\mathbf{n}=\overrightarrow{AB}\times\overrightarrow{AC}=\langle8,7,5\rangle.

Plane: 8(x−1)+7(y−2)+5(z−3)=0⇒8x+7y+5z=37.8(x-1)+7(y-2)+5(z-3)=0\Rightarrow 8x+7y+5z=37.

Worked Example  ·  Angle between two planes

Find the angle between planes π1:2x−y+2z=5\pi_1: 2x-y+2z=5 and π2:x+2y−2z=3\pi_2: x+2y-2z=3.

Solution: n1=(2,−1,2)\mathbf{n}_1=(2,-1,2), n2=(1,2,−2)\mathbf{n}_2=(1,2,-2).

cos⁡θ=∣n1⋅n2∣∣n1∣∣n2∣=∣2−2−4∣3⋅3=49  ⟹  θ≈63.6°\cos\theta=\dfrac{|\mathbf{n}_1\cdot\mathbf{n}_2|}{|\mathbf{n}_1||\mathbf{n}_2|}=\dfrac{|2-2-4|}{3\cdot3}=\dfrac{4}{9}\implies\theta\approx63.6°

Worked Example  ·  Line–plane intersection

Line: r=(1,0,2)+t(1,1,−1)\mathbf{r}=(1,0,2)+t(1,1,-1). Plane: x+2y+z=4x+2y+z=4. Find the intersection.

Solution: Substitute: (1+t)+2t+(2−t)=4  ⟹  2t+3=4  ⟹  t=12(1+t)+2t+(2-t)=4\implies 2t+3=4\implies t=\frac{1}{2}.

Point: (32,12,32)\left(\frac{3}{2},\frac{1}{2},\frac{3}{2}\right).

Triangle ABC with sides c, a, b and angles θABCcabθ
Vector geometry in 3D: vectors AB and AC defining a plane with normal vector n = AB × AC perpendicular to both sides.
⚠ Topic 3 Exam Traps: 3D Vector Lines & Planes
Cross Product Anti-Commutativity: a×b=−(b×a)\mathbf{a} \times \mathbf{b} = -(\mathbf{b} \times \mathbf{a}). Pay strict attention to order when computing normal vectors.

Angle Between Line and Plane: The angle ϕ\phi between line direction d\mathbf{d} and plane normal n\mathbf{n} gives cos⁡ϕ=∣d⋅n∣∣d∣∣n∣\cos\phi = \dfrac{|\mathbf{d}\cdot\mathbf{n}|}{|\mathbf{d}||\mathbf{n}|}. But the angle θ\theta between the line and the plane itself is 90∘−ϕ90^\circ - \phi, so sin⁡θ=∣d⋅n∣∣d∣∣n∣\sin\theta = \dfrac{|\mathbf{d}\cdot\mathbf{n}|}{|\mathbf{d}||\mathbf{n}|}. Using cosine for the line-plane angle is a classic trap.

Skew vs Intersecting Lines: In 3D, two lines that are not parallel might not intersect. You must test whether there is a simultaneous solution for parameter values λ\lambda and μ\mu; if not, the lines are skew.
Practice: Find the distance from point P(3,1,−2)P(3,1,-2) to plane 2x−2y+z=52x-2y+z=5.
d=∣2(3)−2(1)+(−2)−5∣4+4+1=∣6−2−2−5∣3=13d=\dfrac{|2(3)-2(1)+(-2)-5|}{\sqrt{4+4+1}}=\dfrac{|6-2-2-5|}{3}=\dfrac{1}{3}
📝 Chapter Quiz
4

Statistics & Probability (HL)

IB Mathematics HL P1, P2 & P3 AA HL AI HL

HL extends SL statistics with unbiased estimators, confidence intervals, hypothesis testing, and (AI HL) the Poisson distribution.

● Unbiased estimators

μ^=xˉ=∑xin(unbiased for μ).\hat{\mu}=\bar{x}=\frac{\sum x_i}{n}\quad(\text{unbiased for }\mu).S2=∑(xi−xˉ)2n−1(unbiased for σ2;  use sn−1 on GDC).S^2=\frac{\sum(x_i-\bar{x})^2}{n-1}\quad(\text{unbiased for }\sigma^2;\;\text{use }s_{n-1}\text{ on GDC}).

● Confidence interval for mean

xˉ±zα/2⋅σn(σ known),xˉ±tα/2,n−1⋅sn(σ unknown).\bar{x}\pm z_{\alpha/2}\cdot\frac{\sigma}{\sqrt{n}}\quad(\sigma\text{ known}),\qquad\bar{x}\pm t_{\alpha/2,n-1}\cdot\frac{s}{\sqrt{n}}\quad(\sigma\text{ unknown}).

Common critical values: 90% CI → z=1.645z=1.645; 95% CI → z=1.96z=1.96; 99% CI → z=2.576.z=2.576.

⊙ Explorer · Confidence interval for the meaninteractive
margin of error2.058
confidence interval(50.24, 54.36)

This explorer uses the z-based formula xˉ±zα/2⋅s/n\bar x\pm z_{\alpha/2}\cdot s/\sqrt n throughout (a good approximation once nn is reasonably large) rather than switching to the tt-interval Worked Example 4.1 uses for its exact small-sample answer — expect the shaded interval here to be a touch narrower than that example's tt-based one at the same 95%. The curve is the sampling distribution of the mean, width s/ns/\sqrt n — shrink it by raising nn, or widen the shaded region by demanding a higher confidence level; both change the interval for the reason the formula says they should.

● Poisson distribution (AI HL)

X∼Po(λ):P(X=k)=e−λλkk!,μ=σ2=λ.X\sim\text{Po}(\lambda):\quad P(X=k)=\frac{e^{-\lambda}\lambda^k}{k!},\quad\mu=\sigma^2=\lambda.

Conditions: events occur independently at a constant average rate λ\lambda per unit interval.

Poisson as approximation to Binomial: when nn is large, pp is small, and np=λnp=\lambda is moderate.

Worked Example 4.1 — CI

Problem. A sample of n=64n=64 gives xˉ=52.3\bar{x}=52.3 and s=8.4.s=8.4. Construct a 95% CI for μ.\mu.

Use tt-interval (σ\sigma unknown): t0.025, 63≈2.00t_{0.025,\,63}\approx2.00 (from tables or GDC).

CI=52.3±2.00⋅8.464=52.3±2.00(1.05)=52.3±2.1=(50.2,  54.4).CI=52.3\pm2.00\cdot\dfrac{8.4}{\sqrt{64}}=52.3\pm2.00(1.05)=52.3\pm2.1=(50.2,\;54.4).

Worked Example 4.2 — Poisson

Problem. Cars arrive at a toll booth at a rate of 3 per minute. Find: (a) P(X=5)P(X=5); (b) P(X≥2).P(X\ge2).

(a) P(X=5)=e−3⋅355!=e−3⋅243120≈0.1008.P(X=5)=\dfrac{e^{-3}\cdot3^5}{5!}=\dfrac{e^{-3}\cdot243}{120}\approx0.1008.

(b) P(X≥2)=1−P(X=0)−P(X=1)=1−e−3−3e−3=1−4e−3≈1−0.1991=0.8009.P(X\ge2)=1-P(X=0)-P(X=1)=1-e^{-3}-3e^{-3}=1-4e^{-3}\approx1-0.1991=0.8009.

⊙ Explorer · The Poisson distributioninteractive
mean μ = λ3.0
P(X = k)0.1008

Default λ=3\lambda=3, k=5k=5 is the toll-booth rate and outcome from Worked Example 4.2 — the amber bar is exactly that example's answer. Drag k to move the highlighted bar and read off any P(X=k)P(X=k) directly; drag λ\lambda and watch the whole distribution's peak shift and spread out (mean = variance = λ\lambda, so a larger rate means a wider, flatter-looking bell).

Unbiased estimator
s2=∑(xi−xˉ)2n−1 s^2=\dfrac{\sum(x_i-\bar{x})^2}{n-1} estimates σ2 \sigma^2
Hypothesis test
State H0,H1 H_0,H_1 ; compute test stat; compare p p -value to α \alpha
Practice Problem

A machine produces defective items at 2 per hour (Poisson). Find P(exactly 3 defective) in a 2-hour period.

Show solution

λ=4.P(X=3)=e−4⋅433!=64e−46≈0.1954.\lambda=4.\quad P(X=3)=\frac{e^{-4}\cdot4^3}{3!}=\frac{64e^{-4}}{6}\approx0.1954.

Worked Example  ·  Hypothesis test for population mean

A sample of 25 gives xˉ=48.2\bar{x}=48.2, s=6s=6. Test H0:μ=50H_0:\mu=50 vs H1:μ≠50H_1:\mu\neq50 at 5% significance.

Solution: t=48.2−506/25=−1.81.2=−1.5t=\dfrac{48.2-50}{6/\sqrt{25}}=\dfrac{-1.8}{1.2}=-1.5.

Critical value t24,0.025≈2.064t_{24,0.025}\approx2.064. Since ∣−1.5∣<2.064|-1.5|<2.064, do not reject H0H_0.

Worked Example  ·  Confidence interval

Construct a 95% CI for the mean of a normal population with known σ=8\sigma=8 given n=36n=36, xˉ=62\bar{x}=62.

Solution: CI=xˉ±zα/2⋅σn=62±1.96⋅86=62±2.61=(59.4, 64.6)\text{CI} = \bar{x}\pm z_{\alpha/2}\cdot\dfrac{\sigma}{\sqrt{n}} = 62\pm1.96\cdot\dfrac{8}{6} = 62\pm2.61 = (59.4,\,64.6)

Bar chart: Poisson Distribution P(X = k) for λ = 3 with categories: k=0 (5), k=1 (14.9), k=2 (22.4), k=3 (22.4), k=4 (16.8), k=5 (10.1)06111722k=05k=114.9k=222.4k=322.4k=416.8k=510.1Poisson Distribution P(X = k) for λ = 3
Discrete Poisson probability distribution with mean rate λ = 3, illustrating mode at k = 2 and k = 3.
⚠ Topic 4 Exam Traps: Estimators & Hypothesis Testing
Sample Variance Divisor: The unbiased estimator of population variance uses denominator n−1n-1: sn−12=1n−1∑(xi−xˉ)2s_{n-1}^2 = \dfrac{1}{n-1}\sum (x_i - \bar{x})^2. The calculator σx\sigma_x uses nn; always use sxs_x (the n−1n-1 version) in IB HL inference.

Poisson Parameter Scaling: If events occur at rate λ\lambda per hour, the rate for 30 minutes is 0.5λ0.5\lambda, and for 2 hours is 2λ2\lambda. Scale λ\lambda before computing probabilities.

Hypothesis Test Conclusion: Never write 'accept H0H_0'. Write 'There is insufficient evidence at the 5% level to reject H0H_0', followed by context in terms of the problem.
Practice: Two independent samples: n1=10, xˉ1=20, s12=4n_1=10,\,\bar{x}_1=20,\,s_1^2=4 and n2=15, xˉ2=18, s22=5n_2=15,\,\bar{x}_2=18,\,s_2^2=5. Test H0:μ1=μ2H_0:\mu_1=\mu_2 at 10% using a tt-test. State your conclusion.
Pooled sp2=9(4)+14(5)23=10623≈4.61s_p^2=\frac{9(4)+14(5)}{23}=\frac{106}{23}\approx4.61. t=24.61(1/10+1/15)≈20.784≈2.55t=\frac{2}{\sqrt{4.61(1/10+1/15)}}\approx\frac{2}{0.784}\approx2.55. Critical value t23,0.05≈1.714t_{23,0.05}\approx1.714. Reject H0H_0.
📝 Chapter Quiz
5

Calculus (HL)

IB Mathematics HL P1, P2 & P3 AA HL AI HL

HL calculus covers the rigorous theoretical foundations of limits and continuity, advanced integration techniques, differential equations, and Maclaurin series — bridging secondary school and undergraduate pure analysis.

● The 3-Part Definition of Continuity at a Point & Complete Taxonomy of Graph Cases

A function ff is continuous at x=cx = c if and only if all three conditions of the Continuity Checklist are satisfied:

  1. Condition 1: f(c)f(c) is defined (meaning cc is in the domain of ff).
  2. Condition 2: lim⁡x→cf(x)\lim_{x\to c} f(x) exists (meaning left and right limits are finite and agree: lim⁡x→c−f(x)=lim⁡x→c+f(x)=L\lim_{x\to c^-} f(x) = \lim_{x\to c^+} f(x) = L).
  3. Condition 3: lim⁡x→cf(x)=f(c)\lim_{x\to c} f(x) = f(c) (the limiting trend matches the actual output).

Failure of any condition produces a specific discontinuity geometry on the Cartesian plane:

  • Case 1 · Continuous: All 3 conditions hold. The curve passes unbroken through (c,f(c))(c, f(c)).
  • Case 2 · Removable (Hole): Condition 1 fails (undefined at cc), but Condition 2 holds (limit LL exists). The graph has an open ring at (c,L)(c, L).
  • Case 3 · Removable (Displaced Point): Conditions 1 and 2 hold, but Condition 3 fails (f(c)≠Lf(c) \neq L). An open ring sits at (c,L)(c, L) while a solid dot sits at (c,f(c))(c, f(c)).
  • Case 4 · Jump Discontinuity: Condition 2 fails because one-sided limits are finite but unequal (lim⁡x→c−f(x)≠lim⁡x→c+f(x)\lim_{x\to c^-} f(x) \neq \lim_{x\to c^+} f(x)). A step break occurs in the graph.
  • Case 5 · Infinite Discontinuity (Vertical Asymptote): Condition 2 fails because one or both one-sided limits grow without bound to ±∞\pm\infty.
  • Case 6 · Essential / Oscillating Discontinuity: Condition 2 fails because function values oscillate infinitely rapidly between bounds as xx approaches cc.

⊙ Explorer · All cases of graphs of continuityinteractive
input x0.525
output f(x)1.613
lim x→1⁻1.500
lim x→1⁺1.500
Continuity testContinuous at c = 1 ✔

Switch between all 6 continuity graph cases and move x to inspect how the 3-part continuity checklist (f(c) defined, lim x→c exists, lim = f(c)) succeeds or fails.

● Precise (ε, δ) & (ε, M) Definitions of Limits & Rigorous Analysis Proofs

Precise Definition (Limit at a Point): lim⁡x→cf(x)=L\displaystyle\lim_{x \to c} f(x) = L means: for every ε>0\varepsilon > 0, there exists δ>0\delta > 0 such that 0<∣x−c∣<δ  ⟹  ∣f(x)−L∣<ε0 < |x - c| < \delta \implies |f(x) - L| < \varepsilon.

Rigorous Proof (Linear): Prove that lim⁡x→3(4x−5)=7\lim_{x\to 3} (4x - 5) = 7.
Scratchwork: ∣(4x−5)−7∣=4∣x−3∣<ε  ⟺  ∣x−3∣<ε4|(4x - 5) - 7| = 4|x - 3| < \varepsilon \iff |x - 3| < \frac{\varepsilon}{4}.
Proof: Let ε>0\varepsilon > 0. Choose δ=ε4\delta = \frac{\varepsilon}{4}. If 0<∣x−3∣<δ0 < |x - 3| < \delta, then ∣(4x−5)−7∣=4∣x−3∣<4(ε4)=ε|(4x - 5) - 7| = 4|x - 3| < 4\left(\frac{\varepsilon}{4}\right) = \varepsilon. Thus lim⁡x→3(4x−5)=7\lim_{x\to 3}(4x - 5) = 7. ■\blacksquare


Rigorous Proof (Quadratic with Bounding): Prove that lim⁡x→2x2=4\lim_{x\to 2} x^2 = 4.
Scratchwork: ∣x2−4∣=∣x−2∣∣x+2∣|x^2 - 4| = |x - 2||x + 2|. Restrict δ≤1  ⟹  1<x<3  ⟹  ∣x+2∣<5\delta \le 1 \implies 1 < x < 3 \implies |x + 2| < 5. Then ∣x2−4∣<5∣x−2∣|x^2 - 4| < 5|x - 2|. Set 5∣x−2∣<ε  ⟺  ∣x−2∣<ε55|x - 2| < \varepsilon \iff |x - 2| < \frac{\varepsilon}{5}. Choose δ=min⁡(1,ε5)\delta = \min\left(1, \frac{\varepsilon}{5}\right).
Proof: Let ε>0\varepsilon > 0. Set δ=min⁡(1,ε5)\delta = \min\left(1, \frac{\varepsilon}{5}\right). If 0<∣x−2∣<δ0 < |x - 2| < \delta, then ∣x+2∣<5|x + 2| < 5 and ∣x−2∣<ε5|x - 2| < \frac{\varepsilon}{5}. Thus ∣x2−4∣=∣x−2∣∣x+2∣<(ε5)(5)=ε|x^2 - 4| = |x - 2||x + 2| < \left(\frac{\varepsilon}{5}\right)(5) = \varepsilon. Hence lim⁡x→2x2=4\lim_{x\to 2} x^2 = 4. ■\blacksquare


Precise Definition (Limit at Infinity): lim⁡x→∞f(x)=L\displaystyle\lim_{x\to\infty} f(x) = L means: for every ε>0\varepsilon > 0, there exists M>0M > 0 such that x>M  ⟹  ∣f(x)−L∣<εx > M \implies |f(x) - L| < \varepsilon.

Rigorous Proof (Limit at Infinity): Prove that lim⁡x→∞2x+1x+3=2\lim_{x\to\infty} \frac{2x+1}{x+3} = 2.
Scratchwork: ∣2x+1x+3−2∣=∣2x+1−2x−6x+3∣=5x+3<5x<ε  ⟺  x>5ε\left|\frac{2x+1}{x+3} - 2\right| = \left|\frac{2x+1 - 2x - 6}{x+3}\right| = \frac{5}{x+3} < \frac{5}{x} < \varepsilon \iff x > \frac{5}{\varepsilon}.
Proof: Let ε>0\varepsilon > 0. Choose M=5εM = \frac{5}{\varepsilon}. For all x>Mx > M, we have x>5ε  ⟹  5x<εx > \frac{5}{\varepsilon} \implies \frac{5}{x} < \varepsilon. Since x>0x > 0, 5x+3<5x<ε\frac{5}{x+3} < \frac{5}{x} < \varepsilon. Therefore ∣2x+1x+3−2∣=5x+3<ε\left|\frac{2x+1}{x+3} - 2\right| = \frac{5}{x+3} < \varepsilon. Thus lim⁡x→∞2x+1x+3=2\lim_{x\to\infty} \frac{2x+1}{x+3} = 2. ■\blacksquare

⊙ Explorer · Precise (ε, δ) limit definitioninteractive
Max valid δ (ε / 2)0.250
Condition statusValid: curve inside ε-band ✔

For any chosen ε-band (horizontal green corridor), a valid δ must keep the entire curve over (c − δ, c + δ) inside the band. Notice δ ≤ ε / 2 guarantees success for f(x) = 2x − 1.

● Integration techniques

Integration by substitution: ∫f(g(x))g′(x) dx=∫f(u) du\displaystyle\int f(g(x))g'(x)\,dx=\int f(u)\,du where u=g(x).u=g(x).

Integration by parts: ∫u dv=uv−∫v du.\displaystyle\int u\,dv=uv-\int v\,du. (Choose uu by LIATE: Log, Inverse trig, Algebraic, Trig, Exponential.)

Partial fractions: decompose rational integrand, then integrate term by term.

● Differential equations

Separable: dydx=f(x)g(y)⇒∫dyg(y)=∫f(x) dx.\dfrac{dy}{dx}=f(x)g(y)\Rightarrow\displaystyle\int\frac{dy}{g(y)}=\int f(x)\,dx.

Homogeneous: dydx=f(y/x);\dfrac{dy}{dx}=f(y/x); substitute y=vx,  dydx=v+xdvdx.y=vx,\;\dfrac{dy}{dx}=v+x\dfrac{dv}{dx}.

Linear first-order: dydx+P(x)y=Q(x).\dfrac{dy}{dx}+P(x)y=Q(x). Integrating factor: μ=e∫P(x) dx.\mu=e^{\int P(x)\,dx}.

● Maclaurin series (AA HL)

f(x)=f(0)+f′(0)x+f′′(0)2!x2+f′′′(0)3!x3+⋯f(x)=f(0)+f'(0)x+\frac{f''(0)}{2!}x^2+\frac{f'''(0)}{3!}x^3+\cdots

Standard series:

ex=1+x+x22!+x33!+⋯ ,sin⁡x=x−x33!+x55!−⋯ ,ln⁡(1+x)=x−x22+x33−⋯e^x=1+x+\frac{x^2}{2!}+\frac{x^3}{3!}+\cdots,\quad\sin x=x-\frac{x^3}{3!}+\frac{x^5}{5!}-\cdots,\quad\ln(1+x)=x-\frac{x^2}{2}+\frac{x^3}{3}-\cdots

⊙ Explorer · The Maclaurin series for cos xinteractive
f(0.1) exact0.995004
Maclaurin approx.0.995004

n=3n=3 with cos x reproduces Worked Example 5.3 exactly (up to the x4x^4 term). Watch the amber curve hug the blue one near x=0x=0 — and peel away further out. Switch the function to sin x to see the same idea apply to the other standard series above — an odd function needs only odd-power terms, so its approximation improves in different-sized jumps than cos x's.

Worked Example 5.1 — Integration by parts

Problem. Find ∫xe2x dx.\displaystyle\int x e^{2x}\,dx.

Let u=x,  dv=e2x dx⇒du=dx,  v=e2x2.u=x,\;dv=e^{2x}\,dx\Rightarrow du=dx,\;v=\dfrac{e^{2x}}{2}.

∫xe2x dx=xe2x2−∫e2x2 dx=xe2x2−e2x4+C=e2x(2x−1)4+C.\int xe^{2x}\,dx=\dfrac{xe^{2x}}{2}-\int\dfrac{e^{2x}}{2}\,dx=\dfrac{xe^{2x}}{2}-\dfrac{e^{2x}}{4}+C=\dfrac{e^{2x}(2x-1)}{4}+C.

Worked Example 5.2 — Separable DE

Problem. Solve dydx=2xy+1\dfrac{dy}{dx}=\dfrac{2x}{y+1}, given y(0)=2.y(0)=2.

Separate: (y+1) dy=2x dx.(y+1)\,dy=2x\,dx. Integrate: (y+1)22=x2+C.\dfrac{(y+1)^2}{2}=x^2+C.

Initial condition y(0)=2y(0)=2: 92=C.\dfrac{9}{2}=C. Solution: (y+1)2=2x2+9.(y+1)^2=2x^2+9.

Worked Example 5.3 — Maclaurin series

Problem. Find the Maclaurin series for f(x)=cos⁡xf(x)=\cos x up to the x4x^4 term and use it to approximate cos⁡(0.1)\cos(0.1).

cos⁡x=1−x22!+x44!−⋯=1−x22+x424−⋯\cos x=1-\dfrac{x^2}{2!}+\dfrac{x^4}{4!}-\cdots=1-\dfrac{x^2}{2}+\dfrac{x^4}{24}-\cdots

cos⁡(0.1)≈1−0.012+0.000124=1−0.005+0.000004≈0.995004.\cos(0.1)\approx1-\dfrac{0.01}{2}+\dfrac{0.0001}{24}=1-0.005+0.000004\approx0.995004.

Practice Problem

Solve dydx+2xy=x2,  x>0\dfrac{dy}{dx}+\dfrac{2}{x}y=x^2,\;x>0 with y(1)=1.y(1)=1.

Show solution

I.F. μ=x2.ddx(x2y)=x4.\mu=x^2.\quad\frac{d}{dx}(x^2y)=x^4.

x2y=x55+C.y(1)=1⇒C=45.x^2y=\frac{x^5}{5}+C.\quad y(1)=1\Rightarrow C=\frac{4}{5}.

y=x35+45x2.y=\dfrac{x^3}{5}+\dfrac{4}{5x^2}.

Worked Example  ·  Integration by parts

Evaluate ∫x2ex dx\displaystyle\int x^2 e^x\,dx.

Solution: Apply IBP twice. Let u=x2, dv=exdxu=x^2,\,dv=e^x dx:

=x2ex−2∫xex dx=x2ex−2(xex−ex)+C=ex(x2−2x+2)+C= x^2 e^x - 2\displaystyle\int xe^x\,dx = x^2 e^x - 2(xe^x - e^x)+C = e^x(x^2-2x+2)+C

Worked Example  ·  Differential equation

Solve dydx=xy\dfrac{dy}{dx}=\dfrac{x}{y} with y(0)=3y(0)=3.

Solution: Separate: y dy=x dx  ⟹  y22=x22+Cy\,dy = x\,dx\implies\dfrac{y^2}{2}=\dfrac{x^2}{2}+C.

At x=0,y=3x=0,y=3: C=92C=\frac{9}{2}. So y2=x2+9  ⟹  y=x2+9y^2 = x^2+9\implies y=\sqrt{x^2+9}.

Worked Example  ·  Maclaurin series

Write the Maclaurin series for exsin⁡xe^x\sin x up to and including the x4x^4 term.

Solution: ex=1+x+x22+x36+⋯e^x=1+x+\frac{x^2}{2}+\frac{x^3}{6}+\cdots, sin⁡x=x−x36+⋯\sin x=x-\frac{x^3}{6}+\cdots

Product: x+x2+x32−x36+⋯=x+x2+x33−x430+⋯x+x^2+\frac{x^3}{2}-\frac{x^3}{6}+\cdots = x+x^2+\frac{x^3}{3}-\frac{x^4}{30}+\cdots

Coordinate plane with graph of f(x) = x e^(-x); marked points Max (1, 1/e) (1, 0.368), Root (0, 0) (0, 0)1234xyf(x) = x e⁻ˣMax (1, 1/e)Root (0, 0)
Curve f(x) = x e^(-x) integrated using integration by parts, showing local maximum at (1, 1/e) and horizontal asymptote y = 0.
⚠ Topic 5 Exam Traps: Integration & Differential Equations
Constant of Integration in DEs: The constant +C+C must be introduced at the exact moment of integration: ∫1ydy=∫f(x)dx  ⟹  ln⁡∣y∣=F(x)+C\int \frac{1}{y}dy = \int f(x)dx \implies \ln|y| = F(x) + C. Exponentiating gives ∣y∣=eF(x)+C=AeF(x)|y| = e^{F(x)+C} = A e^{F(x)}. Adding +C+C at the very end as y=eF(x)+Cy = e^{F(x)} + C is invalid.

Integration by Parts Sign Error: ∫u dv=uv−∫v du\int u\,dv = uv - \int v\,du. Watch the minus sign when vv itself carries a negative sign (e.g. ∫sin⁡x dx=−cos⁡x\int \sin x\,dx = -\cos x).

Maclaurin Validity: The series expansion for ln⁡(1+x)=x−x22+⋯\ln(1+x) = x - \frac{x^2}{2} + \cdots is only valid for −1<x≤1-1 < x \leq 1.
Practice: Use integration by substitution to find ∫x1−x2 dx\displaystyle\int \dfrac{x}{\sqrt{1-x^2}}\,dx.
Let u=1−x2u=1-x^2, du=−2x dxdu=-2x\,dx. Integral =−12∫u−1/2du=−1−x2+C=-\frac{1}{2}\int u^{-1/2}du=-\sqrt{1-x^2}+C.
📝 Chapter Quiz
AA

AA HL — Additional Topics

IB Mathematics HL — AA P1, P2 & P3 AA HL only

These topics are specific to Analysis and Approaches HL.

● Convergence of series

Geometric series converges iff ∣r∣<1:S∞=a1−r.\text{Geometric series converges iff }|r|<1:\quad S_\infty=\frac{a}{1-r}.Ratio test: if lim⁡n→∞∣an+1an∣<1⇒series converges absolutely.\text{Ratio test: if }\lim_{n\to\infty}\left|\frac{a_{n+1}}{a_n}\right|<1\Rightarrow\text{series converges absolutely}.

● Graphs of inverse trig functions

sin⁡−1:  [−1,1]→[−π/2,π/2],cos⁡−1:  [−1,1]→[0,π],tan⁡−1:  R→(−π/2,π/2).\sin^{-1}:\;[-1,1]\to[-\pi/2,\pi/2],\quad\cos^{-1}:\;[-1,1]\to[0,\pi],\quad\tan^{-1}:\;\mathbb{R}\to(-\pi/2,\pi/2).ddx(sin⁡−1x)=11−x2,ddx(cos⁡−1x)=−11−x2,ddx(tan⁡−1x)=11+x2.\frac{d}{dx}(\sin^{-1}x)=\frac{1}{\sqrt{1-x^2}},\quad\frac{d}{dx}(\cos^{-1}x)=\frac{-1}{\sqrt{1-x^2}},\quad\frac{d}{dx}(\tan^{-1}x)=\frac{1}{1+x^2}.

⊙ Explorer · Inverse trig functions and their derivativesinteractive
f(x)0.524
f′(x)1.155

Blue is the chosen inverse trig function; amber dashed is its derivative, from the formulas above. sin⁻¹ and cos⁻¹ are only defined on [−1, 1], so their derivative curves diverge near the ends of that interval (division by √(1−x²) → 0) — sin⁻¹'s derivative shoots up toward +∞ at both ends (it's 1/√(1−x²), always positive), while cos⁻¹'s shoots down toward −∞ instead (it's the negative of the same expression) — switch to tan⁻¹ to see a derivative that stays bounded everywhere instead, since 1/(1+x²) never blows up.

Worked Example — Limit and convergence

Problem. Determine if ∑n=1∞n2n\displaystyle\sum_{n=1}^{\infty}\frac{n}{2^n} converges.

Ratio test: ∣an+1an∣=n+12n+1⋅2nn=n+12n→12<1.\left|\dfrac{a_{n+1}}{a_n}\right|=\dfrac{n+1}{2^{n+1}}\cdot\dfrac{2^n}{n}=\dfrac{n+1}{2n}\to\dfrac{1}{2}<1. Series converges.

Mathematical induction
Base case; assume n=k n=k ; prove n=k+1 n=k+1
De Moivre's theorem
(cos⁡θ+isin⁡θ)n=cos⁡nθ+isin⁡nθ (\cos\theta+i\sin\theta)^n=\cos n\theta+i\sin n\theta
Euler's form
eiθ=cos⁡θ+isin⁡θ e^{i\theta}=\cos\theta+i\sin\theta
Coordinate plane with graph of f'(x) = 1/(1+x^2); marked points Peak (0, 1) (0, 1), (1, 0.5) (1, 0.5), (-1, 0.5) (-1, 0.5)-3-2-11231xyf'(x) = 1/(1+x²)Peak (0, 1)(1, 0.5)(-1, 0.5)
Derivative curve f'(x) = 1/(1+x^2) of arctan(x), showing peak at (0, 1) and symmetry about the y-axis.
⚠ Topic AA Exam Traps: Convergence, Inverse Trig & Induction
Ratio Test Inconclusive: If lim⁡n→∞∣an+1an∣=1\lim_{n\to\infty} \left|\frac{a_{n+1}}{a_n}\right| = 1, the ratio test provides no conclusion. The series could converge or diverge. You must employ an alternative method such as the comparison or integral test.

Inverse Trig Range Limits: The principal value range of arcsin⁡(x)\arcsin(x) is [−π2,π2][-\frac{\pi}{2}, \frac{\pi}{2}], but arccos⁡(x)\arccos(x) has range [0,π][0, \pi]. Never provide negative output angles for arccos⁡(x)\arccos(x).

Induction Communication Marks: IB markschemes strictly penalize induction proofs that skip the concluding sentence: 'Since P(1)P(1) is true and P(k)  ⟹  P(k+1)P(k) \implies P(k+1), by PMI P(n)P(n) is true for all n∈Z+n \in \mathbb{Z}^+'.
Practice: Prove that cos⁡3θ=4cos⁡3θ−3cos⁡θ\cos 3\theta = 4\cos^3\theta - 3\cos\theta using De Moivre's theorem.
(cos⁡θ+isin⁡θ)3=cos⁡3θ+isin⁡3θ(\cos\theta+i\sin\theta)^3=\cos3\theta+i\sin3\theta. Expand left side using binomial; equate real parts.
Practice: Find the Maclaurin series for ln⁡(1+2x)\ln(1+2x) up to the x3x^3 term, and state its interval of convergence.
Use ln⁡(1+u)=u−u22+u33−⋯\ln(1+u)=u-\frac{u^2}{2}+\frac{u^3}{3}-\cdots with u=2xu=2x: ln⁡(1+2x)=2x−2x2+83x3−⋯\ln(1+2x)=2x-2x^2+\frac{8}{3}x^3-\cdots. Convergence requires −1<2x≤1  ⟹  −12<x≤12-1 < 2x \le 1 \implies -\frac{1}{2} < x \le \frac{1}{2}.
📝 Chapter Quiz
AI

AI HL — Additional Topics

IB Mathematics HL — AI P1, P2 & P3 AI HL only

These topics are specific to Applications and Interpretation HL.

● Matrices

For 2×22\times2 matrix A=(abcd)A=\begin{pmatrix}a&b\\c&d\end{pmatrix}:

det⁡A=ad−bc,A−1=1ad−bc(d−b−ca).\det A=ad-bc,\quad A^{-1}=\frac{1}{ad-bc}\begin{pmatrix}d&-b\\-c&a\end{pmatrix}.

Solving Ax=bA\mathbf{x}=\mathbf{b}: x=A−1b.\mathbf{x}=A^{-1}\mathbf{b}. Eigenvalues: det⁡(A−λI)=0.\det(A-\lambda I)=0.

● Markov chains

A Markov chain models a system that moves between a fixed set of states, where the probability of the next state depends only on the current one (not the history). If pn\mathbf{p}_n is the state-probability (column) vector at step nn and TT is the transition matrix (column jj holds the probabilities of moving from state jj), then:

pn+1=Tpn⇒pn=Tnp0.\mathbf{p}_{n+1}=T\mathbf{p}_n\quad\Rightarrow\quad\mathbf{p}_n=T^n\mathbf{p}_0.

A steady state s\mathbf{s} satisfies Ts=sT\mathbf{s}=\mathbf{s} — it is an eigenvector of TT with eigenvalue 1, normalised so its entries sum to 1. For a well-behaved (regular) chain, pn→s\mathbf{p}_n\to\mathbf{s} as n→∞n\to\infty, regardless of the starting distribution p0\mathbf{p}_0.

⊙ Explorer · Markov chain — approach to steady stateinteractive
steady state πA0.571
P(A) after 15 steps0.571

The transition matrix is T = [[P(A→A), 1−P(B→B)], [1−P(A→A), P(B→B)]]. For a regular chain (not both P(A→A) and P(B→B) at the same extreme), whatever starting P(A) you choose, the curve settles onto the dashed steady-state line πA = (1−P(B→B)) / [(1−P(A→A))+(1−P(B→B))] — an eigenvector of T for eigenvalue 1, exactly as defined above. Try starting P(A) = 0 vs 1 with the same P(A→A), P(B→B): different paths, same destination. Push both sliders to the same edge and that regularity breaks: at P(A→A) = P(B→B) = 1 every state is absorbing, so the chain never moves at all and 'steady state' is just wherever you started; at P(A→A) = P(B→B) = 0 the chain swaps states every single step and never settles — it oscillates between P(A) and 1−P(A) forever.

● Graph theory (networks)

A graph consists of vertices and edges. Weighted graph: edges have assigned weights (distances, costs). Key algorithms:

Kruskal's algorithm — minimum spanning tree (MST): add edges in increasing weight, skip cycles. Prim's algorithm — MST from a vertex; add cheapest edge to tree at each step.

Dijkstra's algorithm — shortest path from a source vertex.

Euler circuit: traverses every edge exactly once; requires all vertices to have even degree.

Hamiltonian path: visits every vertex exactly once.

⊙ Explorer · Kruskal's algorithm, step by stepinteractive
MST weight so far10
edges in tree4 of 4 needed

This is exactly the graph from the Practice Problem below (AB=3, BC=2, CD=4, DE=1, AC=7, BD=5, CE=6, AD=8), sorted by weight: DE, BC, AB, CD, BD, CE, AC, AD. Drag the slider to the right one step at a time — green solid edges were added (they connected two previously-separate pieces of the tree), grey dashed edges were skipped (both endpoints were already connected, so adding one would create a cycle). A spanning tree on 5 vertices always needs exactly 4 edges — once the count reaches 4, every later edge in the list is necessarily skipped, whatever weight it has.

Worked Example — Matrix system

Problem. Solve using matrices: 2x+3y=7,  x−y=1.2x+3y=7,\;x-y=1.

A=(231−1),  b=(71).det⁡A=2(−1)−3(1)=−5.A=\begin{pmatrix}2&3\\1&-1\end{pmatrix},\;\mathbf{b}=\begin{pmatrix}7\\1\end{pmatrix}.\quad\det A=2(-1)-3(1)=-5.

A−1=1−5(−1−3−12)=(0.20.60.2−0.4).A^{-1}=\dfrac{1}{-5}\begin{pmatrix}-1&-3\\-1&2\end{pmatrix}=\begin{pmatrix}0.2&0.6\\0.2&-0.4\end{pmatrix}.

x=A−1b=(0.2(7)+0.6(1)0.2(7)−0.4(1))=(21).x=2,  y=1.\mathbf{x}=A^{-1}\mathbf{b}=\begin{pmatrix}0.2(7)+0.6(1)\\0.2(7)-0.4(1)\end{pmatrix}=\begin{pmatrix}2\\1\end{pmatrix}.\quad x=2,\;y=1.

Markov chains
pn+1=Tpn \mathbf{p}_{n+1}=T\mathbf{p}_n ; steady state: Ts=s T\mathbf{s}=\mathbf{s}
Transition matrix
Each column is a probability distribution from a state
Graph theory (AI HL)
Euler path: all edges once; Hamiltonian: all vertices once
Bar chart: Markov Steady State Distribution (%) with categories: State A (57), State B (43)014294357State A57State B43Probability (%)Markov Steady State Distribution (%)
Markov chain two-state stationary probability distribution showing 57% in state A and 43% in state B.
⚠ Topic AI HL Exam Traps: Transition Matrices & Graph Algorithms
Column-Stochastic vs Row-Stochastic: The IB AI HL syllabus uses column-stochastic transition matrices where each column sums to 1: entry TijT_{ij} represents the probability of moving from state jj to state ii. Multiplying Tpn=pn+1T \mathbf{p}_n = \mathbf{p}_{n+1} requires a column vector. If you set up rows summing to 1 by mistake, standard matrix multiplications produce wrong states.

Cycle Detection in Kruskal's: In Kruskal's algorithm, check for cycles across the entire connected subgraph, not just whether both vertices have been visited. If two visited vertices already belong to the same component, connecting them creates a cycle and the edge must be rejected.

Eulerian Graph Conditions: An Eulerian trail (open path visiting every edge once) requires exactly two vertices of odd degree. An Eulerian circuit (closed loop returning to the start) requires all vertices to have even degree.
Practice: A network has edges: DE=1, BC=2, AB=3, CD=4, BD=5, CE=6, AC=7, AD=8. Use Kruskal's algorithm to determine the minimum spanning tree and its total weight.
Sort edges: DE(1), BC(2), AB(3), CD(4), BD(5)... Add DE, BC, AB, CD without cycles (4 edges connect 5 vertices). Total weight = 1 + 2 + 3 + 4 = 10.
Practice: A network has weighted adjacency matrix WW. Explain how the minimum spanning tree algorithm (Kruskal's) differs from Dijkstra's algorithm.
Kruskal's builds the MST by always adding the globally cheapest edge that does not form a cycle. Dijkstra's finds shortest paths from a single source, greedily updating distance estimates.
📝 Chapter Quiz
§

Practice Set

1Easy

Express z=−1+i3z = -1 + i\sqrt{3} in polar form r(cos⁡θ+isin⁡θ)r(\cos\theta + i\sin\theta) and find z6.z^6.

r=∣z∣=1+3=2,  θ=π−π3=2π3.r=|z|=\sqrt{1+3}=2,\;\theta = \pi - \tfrac{\pi}{3} = \tfrac{2\pi}{3}. So z=2 ⁣(cos⁡2π3+isin⁡2π3).z = 2\!\left(\cos\tfrac{2\pi}{3}+i\sin\tfrac{2\pi}{3}\right).
By De Moivre: z6=26 ⁣(cos⁡4π+isin⁡4π)=64(1+0i)=64.z^6 = 2^6\!\left(\cos 4\pi + i\sin 4\pi\right) = 64(1+0i) = 64.
2Easy

Prove by induction that 5n−15^n - 1 is divisible by 4 for all positive integers n.n.

Base: n=1n=1: 51−1=4.5^1-1=4. ✓
Inductive step: Assume 4∣(5k−1)4\mid(5^k-1), i.e. 5k=4m+1.5^k=4m+1. Then 5k+1−1=5⋅5k−1=5(4m+1)−1=20m+4=4(5m+1).5^{k+1}-1 = 5\cdot5^k-1 = 5(4m+1)-1 = 20m+4 = 4(5m+1). ✓
By induction, 4∣(5n−1)4\mid(5^n-1) for all n∈Z+.n\in\mathbb{Z}^+.
3Easy

Let a=(2−13)\mathbf{a} = \begin{pmatrix}2\\-1\\3\end{pmatrix} and b=(14−2).\mathbf{b} = \begin{pmatrix}1\\4\\-2\end{pmatrix}. Find a×b.\mathbf{a}\times\mathbf{b}.

a×b=∣ijk2−1314−2∣=i[(−1)(−2)−(3)(4)]−j[(2)(−2)−(3)(1)]+k[(2)(4)−(−1)(1)]\mathbf{a}\times\mathbf{b} = \begin{vmatrix}\mathbf{i}&\mathbf{j}&\mathbf{k}\\2&-1&3\\1&4&-2\end{vmatrix} = \mathbf{i}[(-1)(-2)-(3)(4)] - \mathbf{j}[(2)(-2)-(3)(1)] + \mathbf{k}[(2)(4)-(-1)(1)]
=i(2−12)−j(−4−3)+k(8+1)=(−1079).= \mathbf{i}(2-12) - \mathbf{j}(-4-3) + \mathbf{k}(8+1) = \begin{pmatrix}-10\\7\\9\end{pmatrix}.
4Medium

Find ∫01xln⁡(x+1) dx.\displaystyle\int_0^1 x\ln(x+1)\,dx.

Integration by parts: u=ln⁡(x+1),  dv=x dx⇒du=1x+1dx,  v=x22.u=\ln(x+1),\;dv=x\,dx\Rightarrow du=\tfrac{1}{x+1}dx,\;v=\tfrac{x^2}{2}.
=[x22ln⁡(x+1)]01−∫01x22(x+1) dx.=\Big[\tfrac{x^2}{2}\ln(x+1)\Big]_0^1 - \displaystyle\int_0^1\tfrac{x^2}{2(x+1)}\,dx.
First term: 12ln⁡2−0=ln⁡22.\tfrac{1}{2}\ln 2 - 0 = \tfrac{\ln 2}{2}. For the integral, write x2x+1=x−1+1x+1\tfrac{x^2}{x+1}=x-1+\tfrac{1}{x+1}, so ∫01x22(x+1)dx=12[x22−x+ln⁡(x+1)]01=12(12−1+ln⁡2)=ln⁡2−122.\displaystyle\int_0^1\tfrac{x^2}{2(x+1)}dx = \tfrac{1}{2}\Big[\tfrac{x^2}{2}-x+\ln(x+1)\Big]_0^1 = \tfrac{1}{2}(\tfrac{1}{2}-1+\ln2) = \tfrac{\ln2-\frac12}{2}.
Answer: ln⁡22−ln⁡2−122=ln⁡22−ln⁡22+14=14.\tfrac{\ln2}{2} - \tfrac{\ln2-\frac12}{2} = \tfrac{\ln2}{2} - \tfrac{\ln2}{2} + \tfrac{1}{4} = \tfrac{1}{4}.
5Medium

A Poisson random variable X∼Po(3).X\sim\text{Po}(3). Find P(X≥2).P(X\geq 2).

P(X≥2)=1−P(X=0)−P(X=1)=1−e−3−3e−3=1−4e−3≈1−4(0.0498)≈1−0.1992=0.8008.P(X\geq 2) = 1 - P(X=0) - P(X=1) = 1 - e^{-3} - 3e^{-3} = 1 - 4e^{-3}\approx 1 - 4(0.0498) \approx 1 - 0.1992 = 0.8008.
6Medium

Solve the differential equation dydx=2xy2\dfrac{dy}{dx} = 2xy^2 with initial condition y(0)=1.y(0) = 1.

Separate variables: dyy2=2x dx.\dfrac{dy}{y^2} = 2x\,dx. Integrate: −1y=x2+C.-\dfrac{1}{y} = x^2 + C.
Apply IC: −1=0+C⇒C=−1.-1 = 0 + C\Rightarrow C=-1. So −1y=x2−1⇒y=11−x2,  ∣x∣<1.-\dfrac{1}{y} = x^2-1\Rightarrow y = \dfrac{1}{1-x^2},\;|x|<1.
7Medium

Prove that 2\sqrt{2} is irrational.

Proof by contradiction: Suppose 2=pq\sqrt{2} = \dfrac{p}{q} in lowest terms (gcd⁡(p,q)=1\gcd(p,q)=1).
Then 2=p2q2⇒p2=2q2.2 = \dfrac{p^2}{q^2}\Rightarrow p^2=2q^2. So p2p^2 is even, meaning pp is even. Write p=2kp=2k.
Then 4k2=2q2⇒q2=2k24k^2=2q^2\Rightarrow q^2=2k^2, so qq is also even.
This contradicts gcd⁡(p,q)=1.\gcd(p,q)=1. Therefore 2\sqrt{2} is irrational. □\square
8Medium

Find the equation of the plane passing through A(1,0,2),  B(3,1,0),  C(0,2,1).A(1,0,2),\;B(3,1,0),\;C(0,2,1).

AB→=(21−2),  AC→=(−12−1).\overrightarrow{AB}=\begin{pmatrix}2\\1\\-2\end{pmatrix},\;\overrightarrow{AC}=\begin{pmatrix}-1\\2\\-1\end{pmatrix}.
Normal n=AB→×AC→=∣ijk21−2−12−1∣=((−1+4)(2+2)⋅(−1)(4+1))\mathbf{n}=\overrightarrow{AB}\times\overrightarrow{AC}=\begin{vmatrix}\mathbf{i}&\mathbf{j}&\mathbf{k}\\2&1&-2\\-1&2&-1\end{vmatrix}=\begin{pmatrix}(-1+4)\\(2+2)\cdot(-1)\\(4+1)\end{pmatrix}
n=(3−45).\mathbf{n}=\begin{pmatrix}3\\-4\\5\end{pmatrix}. Using point A(1,0,2)A(1,0,2): 3(x−1)−4(y)+5(z−2)=0⇒3x−4y+5z=13.3(x-1)-4(y)+5(z-2)=0\Rightarrow 3x-4y+5z=13.
9Hard

Use the binomial theorem to find the coefficient of x4x^4 in the expansion of (2x−1x)8.(2x - \tfrac{1}{x})^8.

General term: (8r)(2x)8−r ⁣(−1x)r=(8r)28−r(−1)rx8−r−r=(8r)28−r(−1)rx8−2r.\binom{8}{r}(2x)^{8-r}\!\left(-\tfrac{1}{x}\right)^r = \binom{8}{r}2^{8-r}(-1)^r x^{8-r-r} = \binom{8}{r}2^{8-r}(-1)^r x^{8-2r}.
For x4x^4: 8−2r=4⇒r=2.8-2r=4\Rightarrow r=2. Coefficient: (82)⋅26⋅(−1)2=28×64×1=1792.\binom{8}{2}\cdot2^6\cdot(-1)^2 = 28\times64\times1 = 1792.
10Hard

Find all complex roots of z3=−8z^3 = -8 and plot them on the Argand diagram (describe positions).

Write −8=8eiπ.-8 = 8e^{i\pi}. The three cube roots are zk=2ei(π+2πk)/3z_k = 2e^{i(\pi+2\pi k)/3} for k=0,1,2.k=0,1,2.
z0=2eiπ/3=2(12+i32)=1+i3.z_0 = 2e^{i\pi/3} = 2(\tfrac{1}{2}+i\tfrac{\sqrt3}{2}) = 1+i\sqrt3.
z1=2eiπ=−2.z_1 = 2e^{i\pi} = -2.
z2=2ei5π/3=2(12−i32)=1−i3.z_2 = 2e^{i5\pi/3} = 2(\tfrac{1}{2}-i\tfrac{\sqrt3}{2}) = 1-i\sqrt3.
On the Argand diagram, the three roots lie on a circle of radius 2, equally spaced at 120° apart, with z1=−2z_1=-2 on the negative real axis.
11Hard

Solve the second-order ODE y′′−5y′+6y=0y'' - 5y' + 6y = 0 with y(0)=1,  y′(0)=3.y(0)=1,\;y'(0)=3.

Characteristic equation: m2−5m+6=0⇒(m−2)(m−3)=0⇒m=2,3.m^2-5m+6=0\Rightarrow(m-2)(m-3)=0\Rightarrow m=2,3.
General solution: y=Ae2x+Be3x.y=Ae^{2x}+Be^{3x}.
y(0)=1⇒A+B=1.y(0)=1\Rightarrow A+B=1.
y′=2Ae2x+3Be3x,  y′(0)=3⇒2A+3B=3.y'=2Ae^{2x}+3Be^{3x},\;y'(0)=3\Rightarrow 2A+3B=3.
Solving: A=0,  B=1.A=0,\;B=1. So y=e3x.y = e^{3x}.
12Hard

(IB AI HL) A transition matrix T=(0.70.20.30.8)T = \begin{pmatrix}0.7&0.2\\0.3&0.8\end{pmatrix} represents weekly customer movement between two brands. Find the steady-state distribution.

Find eigenvector for eigenvalue λ=1.\lambda=1. Solve (T−I)x=0(T-I)\mathbf{x}=\mathbf{0}:
(−0.30.20.3−0.2)(x1x2)=0⇒−0.3x1+0.2x2=0⇒x2=32x1.\begin{pmatrix}-0.3&0.2\\0.3&-0.2\end{pmatrix}\begin{pmatrix}x_1\\x_2\end{pmatrix}=\mathbf{0}\Rightarrow -0.3x_1+0.2x_2=0\Rightarrow x_2=\tfrac{3}{2}x_1.
Normalise: x1+x2=1⇒x1+32x1=1⇒x1=25,  x2=35.x_1+x_2=1\Rightarrow x_1+\tfrac{3}{2}x_1=1\Rightarrow x_1=\tfrac{2}{5},\;x_2=\tfrac{3}{5}.
Steady-state: Brand A holds 40%, Brand B holds 60% of customers.
✎

Test Generator

Generate a full IB Math HL exam

Creates Paper 1 (no GDC) + Paper 2 (GDC) + Paper 3 (GDC, two extended problem-solving questions) following the IB AA HL format with short-response and extended-response questions.

📝 Chapter Quiz
↓

Downloads & Resources

IB Mathematics HL — Key resources

Paper 3 is the extended investigation paper (60 min). It assesses problem-solving, mathematical communication, and exploration — no specific syllabus topics are assigned to Paper 3.

HL Formula Booklet
More extensive than SL; includes series, complex number forms, vector formulae, and distribution tables.
AA HL Past Papers
Available 2019–present (new syllabus). Focus on proof questions, calculus, complex numbers, and vectors.
AI HL Past Papers
Available 2021–present. Focus on statistics, modelling, matrices, and network problems.
Paper 3 Preparation
Practice with past Paper 3 questions; focus on mathematical communication and justification. Typically 2 questions, each 25–30 marks.
Official · Past Papers

Past Papers & Official Tests

Officially released exams and mark schemes from the examining body.

Worksheet Library

14 original ClipSAT worksheets with full answer keys, organized by unit — free to download and print.

Loading worksheet library…

Mock Exam

--:--