GPA Guarantee: Score A+ under Finals

Lecture Transcription |LMS sync |Live Tutor Q&A
50% OFFAll Plans

ECON1202 Cheatsheet · Quantitative Analysis

UNSW Sydney · Cheatsheet for the final exam. Term 3 exam period, 27 November – 10 December 2026. Full ECON1202 study guide →

A4 landscape · two-side ECON1202 revision · 6 columns per sidev2 · 2026-10-25
ECON1202
Quantitative Analysis
UNSW Sydney · Economics · Level 1 · Term 3 2026
ECON1202 cheatsheet QR code – open the full ECON1202 study guide
Full guide + ask Sia
about ECON1202
SIDE 1/2   DEFINITIONS · FORMULAS · DECISION RULES · CHECKSEXAM REVISION SHEETFinal Exam 60% · priorities before chapter order
Quizzes 10%In-session Tests 20%Project (Excel Assignment) 10%Final Exam 60%
Permitted materials: not confirmed for this offering in the course materials we hold — check the exam instructions on the LMS.Source: course.json + BRIEF · 2026-10-25

1 · Cash-flow ClockCh 1 · final priority

Before calculating, name focal date, rate period and number of those periods. Positive rate + forward time raises value; discounting lowers it. For q nominal compounded k times yearly, r=q/k and n=yk.

single cash flowS=P(1+r)ⁿ  ·  P=S(1+r)⁻ⁿ
continuous: S=Peʲᵗ  ·  unknown n=ln(S/P)/ln(1+r)

effective annual comparisonEAR=(1+q/k)ᵏ−1. Compare offers only on a common horizon; multi-stage value = P × product of each stage factor.

quote q, kyears yr=q/kn=ykfactor(1+r)ⁿ
Match rate and clock. Period rate and exponent must use the same interval.
If askedMoveClose with
Future valuemultiplyS>P if r>0
Present valuedivideP<S if r>0
Target timelogstest adjacent complete periods
Rate quoteone-year factorrank EARs

Translation audit

Given phraseWrite before keys
nominal annual q, monthlyr=q/12; n=12×years
effective annualone yearly factor 1+r
continuous annual jtime t stays in years
rate changes part-wayone factor per regime; multiply

For unknown n, logs give the crossing time; if crediting uses complete periods, verify both neighbouring integers. Across rate changes, multiply separate stage factors rather than averaging rates.

2 · NPV & IRRCh 2 · decision

signed valuationNPV=F₀+F₁/(1+r)+⋯+Fₜ/(1+r)ᵀ
IRR is r solving NPV=0; required return is the external benchmark.

ResultDecision / warning
NPV>0accept at stated required return
NPV<0reject at stated required return
unusual sign changesIRR may have several or no meaningful roots

Equation-of-value method

    3 · Annuity RecognitionCh 3 · timing

    ordinarydue12…n×(1+r)
    First payment decides. End of period = ordinary; immediate = due.

    ordinary annuityPV=R[1−(1+r)⁻ⁿ]/r
    FV=R[(1+r)ⁿ−1]/r

    timing variantsdue value=(1+r)×ordinary value
    loan R=A·r/[1−(1+r)⁻ⁿ]  ·  perpetuity PV=R/r

    StreamRecognitionCheck
    ordinaryequal end paymentsfirst at 1
    dueequal start paymentsfirst at 0
    loanprincipal now; instalments laterPV of payments=A
    mixedlump sum + streamvalue separately, then join

    Never combine at different dates. Rate interval must match payment interval; n counts payments, not merely calendar labels.

    Method: mark time 0, first and last payments and the valuation date; confirm equal amount, spacing and finite count; choose PV or FV from the requested date. Explain a due shift and decompose mixed streams.

    Direction checkExpected relation
    saving, r>0final balance exceeds total contributions
    loan, r>0total instalments exceed principal
    due vs ordinary, same focal datedue value is larger by factor 1+r
    present value of future paymentsbelow their undiscounted total

    4 · Counting ChooserCh 4 · structure

    what differs?order mattersorder ignoredpermutationcombination
    Classify first. Then decide repetition and restrictions.

    core countsⁿPᵣ=n!/(n−r)!  ·  ⁿCᵣ=n!/[r!(n−r)!]
    non-negative x₁+⋯+xₖ=n: C(n+k−1,k−1)
    circle with rotations identical: (n−1)!

    5 · Conditional ProbabilityCh 5 · Bayes

    denominator changesP(A|B)=P(A∩B)/P(B), P(B)>0
    P(A∩B)=P(A)P(B|A)
    independence: P(A∩B)=P(A)P(B)

    A .20Aᶜ .80S .75Sᶜ .25S .10Sᶜ .90S paths.20×.75+.80×.10
    Bayes is path accounting. Multiply down, add evidence paths, divide target path by evidence total.

    Bayes reversalP(A|S)=P(A)P(S|A) / [P(A)P(S|A)+P(Aᶜ)P(S|Aᶜ)]

    TaskVisible workingCheck
    conditionalintersection / conditioned eventdenominator is restricted set
    treemultiply path; add endpointssiblings sum to 1
    Bayesall evidence routes belowposterior direction fits evidence
    independencetest equality numericallynot the same as disjoint

    Tree discipline: label events and complements; multiply down, add endpoints; for Bayes divide the target path by all evidence paths.

    total probabilityP(S)=P(A)P(S|A)+P(Aᶜ)P(S|Aᶜ)

    Disjoint positive-probability events are not independent. Compare a posterior with its prior: evidence more likely under A should raise P(A).

    Bound checkRequirement
    branch splitchildren sum to 1
    terminal / posteriorbetween 0 and 1
    weighted mixturebetween component rates
    independenceconditional equals marginal

    6 · Matrix ProductsCh 6 · dimensions

    Am×nBn×pABm×pinner match; outer survive
    Dimensions explain the product. Each entry is a row–column dot product.

    7 · Eliminate, Then ClassifyCh 7 · systems

    Legal row operations: swap rows; multiply a row by a non-zero constant; replace a row by itself plus a multiple of another. Apply each operation across the augmentation bar.

    echelon augmented matrix[0 … 0 | c], c≠0?yesno solutionno; rank=nuniqueno; rank<ninfinite
    Consistency first. A zero row is harmless unless its augmented entry is non-zero.
    Echelon evidenceConclusionClose
    contradiction rowno solutionstop
    consistent, rank=unknownsuniqueback-substitute
    consistent, rank<unknownsinfinitely manyparameterise free variables

    Sequence: declare variable order; build [A|b]; choose a non-zero pivot or swap; clear below; move right and down to a staircase; then scan bottom-up for contradiction before counting pivots.

    Echelon form: pivots move right, entries below them are zero, and zero rows sit last. Rank is the pivot count; free variables follow only after consistency.

    Audit: operations reach the constant; zero pivots trigger a search; only [0…0|c≠0] contradicts; every free column receives a parameter.

    For applied systems, add non-negativity, integrality or capacity restrictions after the algebraic family. Verify in every original equation.

    8 · Determinant & InverseCh 8 · square only

    2×2 inverseA=[a b; c d], det(A)=ad−bc
    A⁻¹=(1/(ad−bc))[d −b; −c a], only if det(A)≠0

    Row actionEffect on det
    replace Rᵢ by Rᵢ+kRⱼunchanged
    swap two rowssign flips
    scale row by kdet scales by k
    triangular form Uproduct of diagonal, after ledger

    Square A: rank n ⇔ det(A)≠0 ⇔ A⁻¹ exists ⇔ Ax=b has one solution for every b. A zero pivot calls for a swap; singularity requires a missing pivot.

    Choose the cheapest route

    9 · EigenpairsCh 9 · two stages

    reasoning chainAv=λv, v≠0 ⇒ (A−λI)v=0
    non-zero v requires det(A−λI)=0

    Solve the scalar characteristic equation first. For each λ, solve the homogeneous system and choose a non-zero representative. Every non-zero scalar multiple is the same eigendirection.

    Shortcut / checkUse
    triangular Aeigenvalues are diagonal entries
    2×2 rootssum=trace(A); product=det(A)
    candidate paircompare Av with λv componentwise
    repeated λsolve null space; do not infer directions from multiplicity

    Workflow: form A−λI; solve det(A−λI)=0; for each root solve the homogeneous system; choose a non-zero representative; verify Av=λv. Preserve diagonal signs and keep roots separate.

    λAction along eigenline
    λ>1stretch, same direction
    0<λ<1shrink, same direction
    λ<0reverse direction and scale
    λ=0map that non-zero vector to zero

    The zero vector is never an eigenvector; non-zero scalar multiples represent the same direction.

    10 · Differentiate by StructureCh 10 · rule choice

    inspect expressionpowerproductquot.chainseveral rules may be nested
    Structure selects the rule. Simplify validly, then mark moving factors and inner functions.

    rules to print(fg)′=f′g+fg′
    (f/g)′=(f′g−g′f)/g², g≠0
    d[f(g(x))]/dx=f′(g(x))g′(x)

    basic derivativesd(xⁿ)/dx=nxⁿ⁻¹  ·  d(eᵍ)/dx=eᵍg′
    d(ln g)/dx=g′/g, with g>0 in the real logarithm used here

    point elasticityη=(p/q)(dq/dp). It is unit-free; sign gives direction, |η| gives local responsiveness at the stated point.

    Failure signalRepair
    product derivative has one termdifferentiate each factor in turn

    Answer path

    11 · Integrate & Model GrowthCh 11 · verify twice

    antiderivatives∫xⁿdx=xⁿ⁺¹/(n+1)+C, n≠−1
    ∫1/x dx=ln|x|+C
    ∫eᵃˣdx=eᵃˣ / a + C, a≠0
    ∫ₐᵇf(x)dx=F(b)−F(a)

    QuestionMust showCheck
    indefinitefamily + Cdifferentiate
    definiteF(b)−F(a), bracket lower valuesign + accumulated units
    total areasplit at zeros if neededadd magnitudes

    differential equationsdy/dt=ky ⇒ y=y₀eᵏᵗ
    dy/dt=k(M−y) ⇒ y=M−(M−y₀)e⁻ᵏᵗ

    y′=kyy′=k(M−y)Mfeedback?
    Different feedback, different shape. Current stock accelerates; the remaining gap shrinks toward M.

    Verify twice: substitute t=0 to recover y₀; differentiate to recover the rate law. With k>0, limited growth approaches M rather than reaching it in finite time.

    Model cueConstant / unitsBehaviour
    rate proportional to levelk has reciprocal-time unitspercentage pattern
    rate proportional to M−yM−y₀ sets initial gapslows near capacity
    proposed solutionsame time unit in ktcheck initial + derivative

    12 · OptimisationCh 12 · classify

    unconstrainedsolve fₓ=0 and fᵧ=0.
    D₁=fₓₓ; D₂=fₓₓfᵧᵧ−fₓᵧ²

    D₂D₁Local class
    >0<0maximum
    >0>0minimum
    <0anysaddle
    =0anytest inconclusive
    candidateg=cconstraint stays active
    Constraint and contour meet. Lagrange conditions locate candidates while enforcing feasibility.
    https://www.asksia.ai/cheatsheet/unsw-econ1202 · Full ECON1202 study guide →
    Shared by ________
    AskSiaForward this to your tutorial group · v2 · 2026-10-25
    Term 3 exam period, 27 November – 10 December 2026.
    Check your exam instructions on the LMS for what you may bring; use this sheet to revise and self-test.
    ECON1202
    Quantitative Analysis
    UNSW Sydney · Worked drills · traps · exam-morning plan
    ECON1202 cheatsheet QR code – open the full ECON1202 study guide
    Full guide + ask Sia
    about ECON1202
    SIDE 2/2   FIVE WORKED DRILLS · IF-X-THEN-Y CARDS · CLASSIC TRAPS · EXAM-MORNING PLANSELF-TEST, THEN COVER ANSWERSKeep intermediate values unrounded

    13 · Drill 1: Mixed Accountcourse worked example

    Q. Start with $2,000; deposit $180 at each month end for 18 months; 4.8% nominal compounded monthly. Balance immediately after the final deposit?

    $2k$180$180…$180+?018r=.048/12=.004
    One focal date. Accumulate the lump sum and ordinary stream separately.
    1. r=0.048/12=0.004; n=18 monthly periods.
    2. Lump sum: 2000(1.004)¹⁸=$2,149.00.
    3. Annuity: 180[(1.004)¹⁸−1]/0.004=$3,352.55.
    4. Same date: 2,149.00+3,352.55=$5,501.55.
    Self-checkEvidence
    timingmonth-end deposits → ordinary
    scaleresult exceeds $2,000+18($180)
    roundingretain factors until cents

    14 · Finance Judgement DeckCh 1–3 self-test

    date + signevery flowmove to onefocal dateNPV / IRRcompare
    Decision sequence. The final sign matters only after all dated amounts share a valuation basis.
    If you seeWrite firstClose with
    nominal compounded k timesr=q/k; n=ykrate and count share a clock
    unknown timeisolate factor; take logstest neighbouring complete periods
    rate comparisonone-year growth factorsrank effective returns
    changing ratesseparate stage factorsmultiply; never average
    unequal dated receiptsone discounted term per flowrebuild the timeline

    NPV versus IRR

    NPV asksIRR asks
    value left at a chosen required returnwhich rate makes NPV zero
    benchmark is an inputrate comes from the cash-flow stream
    sign supplies the stated decisioncompare admissible root with benchmark

    Cover the right column and recite the move. Payment-plan totals cannot be compared before timing is adjusted. A positive future receipt discounts below its nominal amount when r>0. Unusual sign patterns can create multiple or missing meaningful IRRs, so the defining equation and a root screen remain visible.

    Annuity boundary

    15 · Drill 2: BayesCh 5 worked

    Q. 20% of items come from A. Signal rates: 75% for A, 10% otherwise. Find P(A|signal).

    A .20Aᶜ .80S .75Sᶜ .25S .10Sᶜ .90.15+.08=.23
    Circle the observed endpoints. Their sum is the denominator.
    1. Target joint path: 0.20×0.75=0.15.
    2. All signal paths: 0.15+0.80×0.10=0.23.
    3. P(A|S)=0.15/0.23=0.6522 = 65.22%.
    4. Direction: signal lifts the 20% prior because it is more likely under A.
    If you seeDo
    “given B”restrict denominator to B
    reverse conditioningdraw causes → evidence
    “independent?”test joint = product; do not guess
    weighted mixturecheck it lies between component rates

    Cover-and-recite checks

    PromptRequired sentence
    P(A|B)“among B cases, what share also lie in A?”
    P(A∩B)“reach A, then follow the B branch”
    P(B)“add every mutually exclusive route ending in B”
    P(A|B) by Bayes“target A-and-B path divided by all B paths”

    Write event labels beside every decimal. The probability on a second-stage branch is conditional on its parent. Sibling branches leaving one node sum to 1; terminal path weights need not. A posterior must stay in [0,1], and its movement away from the prior should match which cause makes the evidence more likely.

    Independence test: compare P(A∩B) with P(A)P(B), or P(B|A) with P(B). Equal conditional shares support independence even when group sizes differ. Disjoint positive-probability events are dependent because observing one makes the other impossible.

    16 · Counting Micro-testCh 4 method

    Named jobs, ranks and positions make order matter; a committee depends only on membership. Use the swap test before any factorial.

    StructureEngineWhy
    sequential slotsmultiply stage countsone choice from every stage
    ordered selectionⁿPᵣswapping creates a new outcome
    unordered subsetⁿCᵣinternal orders are duplicates

    17 · Drill 3: Matrix MeaningCh 6 worked

    Q. Quantity rows [4,2,5], [3,6,1]; price vector [8,3,4]ᵀ. Find each store total.

    4 2 53 6 18345846(2×3)(3×1) → 2×1
    Rows survive. Each store row meets the common price column.
    1. Dimensions: (2×3)(3×1) ⇒ 2×1.
    2. Store 1: 4(8)+2(3)+5(4)=58.
    3. Store 2: 3(8)+6(3)+1(4)=46.
    4. Answer: [$58,$46]ᵀ, in original store order.
    JudgementReason
    AB exists?inner dimensions match
    BA too?recheck; never infer from AB
    entry (i,j)row i · column j only
    applied outputrestore row labels and units

    Matrix error audit

    ClaimTest before accepting
    sum A + Bsame dimensions; entrywise operation
    product ABA columns = B rows; inner labels align
    rearranged productonly parentheses moved, never factor order
    transposed productorder reversed: (AB)ᵀ=BᵀAᵀ
    system Ax=bunknown order fixed; zero coefficients retained

    Dimensions alone do not guarantee meaning: inner category labels must align. Expand one dot product and restore units.

    18 · Drill 4: InverseCh 8 worked

    A=[3 1;1 2], b=[10,8]ᵀ. det(A)=3(2)−1(1)=5, so A⁻¹=(1/5)[2 −1;−1 3].

    x=A⁻¹b=(1/5)[20−8; −10+24]=[2.4,2.8]ᵀ

    Verify: 3(2.4)+2.8=10; 2.4+2(2.8)=8. The check returns the original b.

    Inverse checkpointEvidence
    shapeA is square

    18b · Row-Reduction JudgementCh 7 method

    echelon [A|b]contradiction [0…0|c≠0]?yes → noneno: r=n → 1no: r<n → ∞
    Read bottom-up. Consistency comes before pivot count and parameterisation.

    19 · Drill 5: Chain RuleCh 10 worked

    Q. f(x)=e^(x²−3x). Find f′ and f′(2).

    inner gx²−3xg′2x−3outereᵍ×g′outer derivative × inner derivative
    Expose the inner function. Its derivative must survive as a factor.
    1. g=x²−3x, so g′=2x−3.
    2. f′(x)=(2x−3)e^(x²−3x).
    3. At x=2: (4−3)e^(4−6)=e⁻²=0.1353 approximately.
    If you seeDoAudit
    u·vu′v+uv′two contributions
    u/v(u′v−v′u)/v²sign + domain
    outer(inner)outer′ × inner′inner factor present
    elasticity(p/q)dq/dpsame point; unit-free

    Derivative answer audit

    Preserve the domain; label the expression parts; write the symbolic rule; differentiate generally; evaluate last and attach units or elasticity meaning.

    Products need both contributions; quotients preserve numerator-prime × denominator minus denominator-prime × numerator; chains retain the inner derivative.

    QuantityInterpretationCheck
    f′(x)local output change per input unitnearby secant has similar sign/size
    η=(p/q)dq/dplocal proportional responseunit-free; same evaluation point
    downward demandquantity falls as price risesη usually negative

    20 · Growth Judgement CardCh 11

    kyk(M−y)Mfeedback?
    Read the right-hand side. Current stock accelerates; remaining gap slows near M.
    Rate lawSolution cueLong run
    y′=kyy₀eᵏᵗgrowth if k>0; decay if k<0
    y′=k(M−y)M−(M−y₀)e⁻ᵏᵗapproaches M if k>0

    Integration self-test

    PromptAnswer must containFast check
    indefinite integralantiderivative family + Cdifferentiate it
    definite net accumulationF(b)−F(a), lower value bracketedsign follows net area
    geometric total areasplit where sign changesadd positive magnitudes

    Differential-equation verification

    21 · Classic Trapsall chapters

    TrapWhat failedRepair
    annual q with monthly nclocks differr=q/12 and count months
    ordinary treated as duefirst date hiddendraw payment at 0 or 1
    P(A|B)=P(B|A)denominator reversedname event after bar
    combination for named jobsorder discardedswap test
    AB changed to BAfactor order lostwrite dimensions + labels
    zero row ⇒ no solutionconstant ignoredlook for [0…0|c≠0]
    det=0 ⇒ no solutionconsistency unknownreduce augmented matrix
    v=0 reportedeigen definition brokenchoose non-zero vector
    stationary ⇒ maximumno classificationD₂ test + boundaries
    ∫ without Cfamily omittedadd C if indefinite
    square: rank n ⇔ det≠0 ⇔ inverse⇔ unique Ax=b for every brank<n ⇔ det=0 ⇔ singularnone vs infinite needs consistency
    Do not skip the boundary. Singularity alone does not choose between none and infinitely many solutions.

    Convert every trap into a check: one clock; correct conditional denominator; dimensions plus labels; consistency before rank; differentiation or substitution back; feasibility plus classification. Record the first incorrect line and the check that would have caught it.

    Determinant ledger

    During eliminationRemember
    row replacementdet unchanged
    one row swapdet sign flips
    row scaled by kdet scaled by k
    triangular finishdiagonal product belongs to transformed matrix until ledger reversed

    22 · Eigenpair AuditCh 9

    vAv=λvsame line; non-zero v
    Fast check. Compute Av and λv separately; equality verifies scale and direction together.
    λ sign / sizeGeometric meaning
    λ>1same direction, stretch
    0<λ<1same direction, shrink
    λ<0reverse along the same line
    λ=0maps eigenvector to zero

    23 · Method Chooserfast recall

    Prompt signalFirst moveFinal check
    dated moneydraw focal date; match r,ndirection + scale
    equal paymentslocate first paymentordinary/due shift
    ordered / selectedswap test; repetition?integer + bound
    given / posteriordefine events; treebounds + evidence direction
    matrix productwrite dimensions + labelsouter size + units
    linear systemaugment; row-reduceconsistency before rank
    inverse / detconfirm squareAA⁻¹=I or Ax=b
    eigenpaircharacteristic equationAv=λv, v≠0
    derivativeinspect expression treedomain + nearby slope
    accumulationantiderivative; bounds?differentiate
    growth lawcurrent stock or gap?initial value + derivative
    optimumFOCs; list candidatesclassification + feasibility
    candidateg=c∇f = λ∇g
    Constraint stays visible. Solve all Lagrange conditions, then recheck feasibility and compare values.

    24 · Exam-Morning Plancourse strategy

    1. Recite the decision boundaries: ordinary/due; permutation/combination; conditional/reversed; AB/BA; redundant/contradictory; stationary/classified.
    2. Warm up once per family: one timing translation, one Bayes denominator, one dot product, one row classification, one chain rule, one Hessian branch.
    3. Use a four-line answer path: define structure → formula/rule → substitution → interpretation + check.
    4. Keep precision: retain intermediate calculator values; round only at the requested final precision.
    5. Run the audit: direction, probability bounds, dimensions, original equations, derivative, initial condition or constraint.
    6. Check current instructions: duration, sections, date, calculator/tool conditions and permitted materials on the LMS.

    Four-box error ledger

    BoxQuestion after each practice item
    model choiceDid timing, event, dimension or constraint select the right method?
    https://www.asksia.ai/cheatsheet/unsw-econ1202 · Full ECON1202 study guide →
    Shared by ________
    AskSiaForward this to your tutorial group · v2 · 2026-10-25
    Term 3 exam period, 27 November – 10 December 2026.
    Check your exam instructions on the LMS for what you may bring; use this sheet to revise and self-test.

    Want one for YOUR exact syllabus?

    Sia is your free desktop study agent. Drop your UNSW Sydney ECON1202 slides — Sia builds a sheet tailored to YOUR exam. Better than this library because it knows YOUR materials.

    ↓ Download Sia · Free