ETX5900 Business Statistics
The Complete Exam Bible · S2 2026ETX5900 Overview
- Monash University
- S2 2026
- 12 chapters
- Statistics
ETX5900 Business Statistics is the postgraduate business-statistics unit taught by the Department of Econometrics and Business Statistics in Monash University's Faculty of Business and Economics.
- Assessed by In-class activities / Exercise (Weeks 2-12, best 9 of 11, group) 15% + 3 more
- Key terms Population vs sample, Standard deviation (s), Standard error, Central Limit Theorem (CLT)
- How to prepare Treat ETX5900 as a method-selection unit, not a memorisation unit - the exam provides the formula sheet and statistical tables.
- Most asked Where can I find past exam papers or practice for ETX5900?
How ETX5900 is assessed
| Component | Weight | Format |
|---|---|---|
| In-class activities / Exercise (Weeks 2-12, best 9 of 11, group) | 15% | In-workshop activities |
| Take-home Quizzes / Quiz-Test (Weeks 1-11, best 10 of 11, 1 attempt, 60 min) | 15% | Individual online quiz |
| Assignment (Written) Part A + Part B on supplied Excel dataset | 20% | Individual data-analysis report + quiz |
| Final Examination | 50% | Individual invigilated e-exam; formula sheet + stat tables provided; duration To be advised |
What ETX5900 covers
ETX5900 Business Statistics builds in one direction: describe data, then quantify uncertainty, estimate, and finally test and model it.
These twelve chapters follow the Monash teaching modules from descriptive statistics and probability, through the normal and sampling distributions and confidence intervals, into the chi-square test, hypothesis testing, correlation and regression, and time-series forecasting - each mapped to its module and to a slice of the 50% final e-exam.
Descriptive Statistics I: Concepts, Data Types & Visualisation
Module 1 / Week 1 - population vs sample, variable types, charts (Berenson Ch 1-2)This is the Week 1 foundation of ETX5900 Business Statistics at Monash University, where the whole unit's vocabulary is set: the difference between a population and a sample, between a parameter (μ, σ, π) and a statistic (x̄, s, p), and how to name each variable's type and level of measurement.
- Descriptive vs inferential statistics: summarising data now vs generalising to a population later
- Population (size N) vs sample (size n): who is measured, and why n ≤ N
- Parameter vs statistic: Greek μ, σ, π describe a population; x̄, s, p come from a sample
- Sampling designs: simple random, systematic, stratified, cluster vs biased convenience/judgement samples
- Categorical data (nominal vs ordinal) vs numerical data (discrete vs continuous)
- The four levels of measurement: nominal < ordinal < interval < ratio, and why a true zero matters
Descriptive Statistics II: Distribution, Summary Measures & Pivot Tables
Module 2 / Week 2 - mean, median, variance, s, CV, pivot tables, data mining (Berenson Ch 3)This is Module 2 (Week 2) of ETX5900 Business Statistics at Monash University, the numerical half of descriptive statistics (Berenson Ch 3).
- Central tendency: mean x̄, median and mode — and when to use each
- The mean-median gap as a fast read of distribution shape
- Measures of variation: range, interquartile range (IQR = Q₃ − Q₁)
- Sample variance s² and standard deviation s — the n−1 (Bessel) divisor vs population N
- Coefficient of variation CV = (s / x̄) × 100% for comparing relative spread across differently-scaled series
- Distribution shape: symmetric vs left/right skew; the empirical 68–95–99.7 rule (bell-shaped data only)
Probability: Concepts & Applications
Module 3 / Week 3 - addition/conditional rules, independence (Berenson Ch 4)Module 3 (Week 3) of ETX5900 Business Statistics at Monash University introduces probability — the arithmetic of uncertainty that underpins every later inference topic in the unit.
- Sample space, events, and the axioms: P(A) in [0,1], P(S) = 1
- Marginal, joint and conditional probability read from a contingency table
- The addition rule: P(A or B) = P(A) + P(B) - P(A and B)
- Mutually exclusive events: no overlap, so P(A and B) = 0
- The complement rule: P(A') = 1 - P(A), the fast route to 'at least one'
- Conditional probability: P(A | B) = P(A and B) / P(B)
The Normal Distribution & Sampling Distributions
Module 4 / Week 4 - Z-scores, CLT, standard error, foundations of inference (Berenson Ch 6-7)This is Module 4 (Week 4) of ETX5900 Business Statistics at Monash University, the postgraduate business-statistics unit run by the Department of Econometrics & Business Statistics (textbook: Berenson et al., Chapters 6-7).
- The normal model X ~ N(μ, σ²) and standardising to Z = (X − μ)/σ ~ N(0,1)
- Reading the cumulative standard-normal table: left = Φ(z), right = 1 − Φ(z), between = Φ(b) − Φ(a)
- The empirical 68-95-99.7 rule and key z landmarks (1.645, 1.960, 2.576; 90th percentile z = 1.28)
- Inverse-normal problems: finding a value or percentile from an area via X = μ + zσ
- The sampling distribution of the sample mean x̄: mean = μ, standard error = σ/√n
- The Central Limit Theorem: x̄ is approximately normal for n ≥ 30 whatever the population shape
Confidence Interval Estimation: Mean & Proportion
Module 5 / Week 5 - z-CI (sigma known), t-CI (sigma unknown), proportion CI (Berenson Ch 8)Confidence Interval Estimation is Week 5 of ETX5900 Business Statistics at Monash University, the first inference topic in the unit (Berenson Ch 8).
- Point estimate vs interval estimate; the universal estimate ± margin-of-error shape
- Confidence level (1−α) and what “95% confident” really means
- z-interval for a mean when σ is KNOWN: x̄ ± z_{α/2}·σ/√n
- t-interval for a mean when σ is UNKNOWN: x̄ ± t_{n−1,α/2}·s/√n, df = n−1
- z-interval for a proportion: p ± z_{α/2}·√(p(1−p)/n)
- Margin of error and how it responds to confidence level, n and σ/s
Chi-Square Test for Independence
Module 6 / Week 6 - categorical vs categorical, expected freqs, df=(r-1)(c-1) (Berenson Ch 15, sec 15.3)Module 6 (Week 6) of ETX5900 Business Statistics at Monash University introduces the chi-square (χ²) test of independence — the formal hypothesis test for whether two categorical variables are related.
- When to use it: two <b>categorical</b> variables cross-classified as counts in an <i>r</i>×<i>c</i> contingency table
- Hypotheses: H₀ = the variables are <b>independent</b> (no relationship); H₁ = they are <b>dependent</b>
- Expected frequency: fₑ = (row total × column total) / n, computed for every cell assuming H₀ is true
- Test statistic: χ² = Σ (fₒ − fₑ)² / fₑ, summed over all cells (always ≥ 0)
- Degrees of freedom: df = (r − 1)(c − 1) — set by the table size, not the sample size
- Validity condition: every expected count fₑ ≥ 5; if not, combine adjacent categories and recompute
Hypothesis Testing for Business Decisions I
Module 7 / Week 7 - five-step framework, one-sample mean test, Type I/II errors (Berenson Ch 9)Module 7 (Week 7) of ETX5900 Business Statistics at Monash University introduces formal hypothesis testing — the procedure that turns a business claim about a population into a decision under uncertainty.
- The five-step framework: state H0 and H1, compute the test statistic, find the critical value or p-value at level alpha, decide, conclude in context
- H0 (null) always contains the equality; H1 (alternative) carries the research claim and sets the tail
- One-sample mean, sigma KNOWN: Z = (xbar - mu0) / (sigma / sqrt(n)), compared to the standard normal
- One-sample mean, sigma UNKNOWN: t = (xbar - mu0) / (s / sqrt(n)) with df = n - 1, compared to Student-t
- Two-tailed (H1: mu != mu0): reject if |statistic| > critical value at alpha/2 in each tail
- One-tailed (H1: mu > mu0 or mu < mu0): reject if the statistic falls in the single tail beyond the alpha critical value
Hypothesis Testing for Business Decisions II
Module 8 / Week 8 - proportion tests, two-sample extensions, R intro (Berenson Ch 9-10)Module 8 (Week 8) of ETX5900 Business Statistics at Monash University extends the hypothesis-testing frame from a single mean to a single proportion and to two-sample comparisons, and introduces R as an analysis tool.
- The five-step testing frame: hypotheses, test statistic, critical value / p-value, decision, conclusion in context
- H0 always contains '='; H1 is the research claim and sets the tail (one- vs two-tailed)
- One-sample proportion z-test: Z = (p - pi0) / sqrt(pi0(1 - pi0) / n), with the SE built from the hypothesised pi0
- Validity condition for the normal approximation: n*pi0 >= 5 and n(1 - pi0) >= 5
- Reject rules by tail: |z| > z(alpha/2) two-tailed, z beyond z(alpha) one-tailed, or p-value < alpha
- Two-sample difference of proportions: pooled p-bar = (X1 + X2)/(n1 + n2) in a z-statistic
Introduction to Correlation & Regression Analysis
Module 9 / Week 9 - Pearson r, least-squares line, b0/b1 (Berenson Ch 12)Module 9 (Week 9) of ETX5900 Business Statistics at Monash University is the unit's first look at modelling the relationship between two numeric variables.
- Scatterplots and the population model Y = beta0 + beta1 X + epsilon vs the fitted line Y-hat = b0 + b1 X
- Pearson correlation r in [-1, +1]: strength and direction of a straight-line relationship
- Correlation is not causation, and r only detects LINEAR patterns
- Least-squares slope b1 = Sxy / Sxx and intercept b0 = y-bar - b1 x-bar
- Interpreting the slope in context (change in Y per one-unit change in X)
- Sums of squares: SST = SSR + SSE, and the coefficient of determination R^2 = SSR/SST = r^2
Regression Model Analysis
Module 10 / Week 10 - slope t-test, R2, SST/SSR/SSE, prediction (Berenson Ch 13)Regression Model Analysis is Week 10 of ETX5900 Business Statistics at Monash University (Berenson Ch 13), the inference half of simple linear regression.
- The estimated line Ŷ = b₀ + b₁X and the least-squares slope b₁ = Σ(x−x̄)(y−ȳ)/Σ(x−x̄)²
- Partitioning variation: SST = SSR + SSE (total = explained + residual)
- Coefficient of determination R² = SSR/SST = 1 − SSE/SST, and R² = r² in simple regression
- The slope t-test H₀: β₁ = 0 vs H₁: β₁ ≠ 0, two-tailed, testing whether a linear relationship exists
- Test statistic t = b₁/se(b₁) with df = n − 2, and se(b₁) = √(MSE/Σ(x−x̄)²), MSE = SSE/(n−2)
- Reject rule: reject H₀ if |t| > t_{n−2,α/2} (equivalently p-value < α)
Time Series Analysis & Forecasting
Module 11 / Week 11 - linear/quadratic trend, MAD, MSFE, multiplicative model (Berenson Ch 14)Time Series Analysis & Forecasting is Module 11 / Week 11 of ETX5900 Business Statistics at Monash University, the unit's final topic (Berenson Ch 14).
- Time series vs an ordinary sample: order matters because value depends on time
- The four components: Trend, Seasonal, Cyclical, Irregular (T, S, C, I)
- Linear trend model Ŷ_t = b₀ + b₁t, with t = 1, 2, 3, … as the time index
- Quadratic trend model Ŷ_t = b₀ + b₁t + b₂t² for accelerating/decelerating series
- Fitting the trend line by least squares (same b₀, b₁ mechanics as regression, with t as predictor)
- Forecasting by substituting the next t into the fitted equation
Past Exams & Revision
Module 12 / Week 12 - mock e-exam, Part A/B/C practice, formula-sheet drillPast Exams & Revision is Week 12 of ETX5900 Business Statistics at Monash University — the synthesis week that turns the whole unit into an exam-ready routine rather than adding new theory.
- The exam blueprint: final individual invigilated e-exam worth 50%, sat in the ~November 2026 end-of-semester period (formula sheet + statistical tables provided; duration to be advised — confirm on Moodle)
- The Part A / B / C structure of the past papers and how marks are distributed across concepts, short calculations and extended data analysis
- The which-procedure decision map: estimation vs testing; mean vs proportion vs categorical vs numeric relationship
- The σ-known (z) vs σ-unknown (t, df = n−1) fork for a mean, and the two different standard errors for a proportion (CI uses p, test uses π₀)
- The five-step hypothesis-testing routine and the correct tail + reject rule for each test
- Reject rules stated correctly: two-tailed |stat| > crit_{α/2}, one-tailed single tail, chi-square upper-tail only, and the universal p < α rule
It builds the full inference toolkit a business analyst needs: describing data, quantifying uncertainty with probability and the normal and sampling distributions, estimating with confidence intervals, and then testing and modelling with the chi-square test, hypothesis testing, correlation and regression, and time-series forecasting.
The unit runs as hands-on weekly workshops built around Excel (with R introduced late in the semester), and it assesses one skill above all at Monash University: taking a business question, choosing the correct statistical procedure, and defending the decision.
Your mark comes from four tasks (in-class activities 15%, take-home quizzes 15%, a written assignment 20%, and the 50% final e-exam); a formula sheet and statistical tables are provided in the exam, so ETX5900 rewards judgement over memorisation.
This guide mirrors the unit as taught, maps every chapter to its module, and uses the SWOTVAC run-in to rehearse method selection before the ~November 2026 exam period (confirm the exact date on Moodle / my.Monash). ETX5900 is a standard Monash postgraduate unit — confirm its credit-point value on Moodle / my.Monash so you can plan your workload across the semester.
Worked example · free
Two-tailed hypothesis test for a mean (sigma unknown, t-test)
- +1Step 1 - Hypotheses. Because the claim is about a difference in either direction, this is a two-tailed test: H0: mu = 45 versus H1: mu is not equal to 45, at significance level alpha = 0.05.
- +1Step 2 - Choose the statistic. The population sigma is unknown and is estimated by the sample s, so use the t-test with df = n - 1 = 19 (use t, not z, whenever sigma is unknown).
- +1Step 3 - Compute t. Standard error = s / √n = 6 / √20 = 6 / 4.472 = 1.342, so t = (x̄ − μ₀) / (s / √n) = (48 − 45) / 1.342 = 3 / 1.342 = 2.236.
- +1Step 4 - Critical value. Two-tailed at alpha = 0.05 puts alpha/2 = 0.025 in each tail; from the t table at df = 19 the critical values are plus or minus 2.093.
- +1Step 5 - Decision. Since |t| = 2.236 is greater than 2.093, the statistic falls in the upper rejection tail, so reject H0 (equivalently the two-tailed p-value is about 0.037, which is less than 0.05).
- +1Step 6 - Conclusion in context. There is sufficient evidence at the 5% level that the mean fill weight differs from 45 g; the sample points to over-filling. Note we say reject H0, never that H0 is proven true.
Key terms
- Population vs sample
- The population is the whole group of interest; a sample is the subset you actually measure. A numerical summary of a population is a parameter (e.g. mu, sigma); of a sample it is a statistic (e.g. x-bar, s).
- Standard deviation (s)
- A measure of spread: s = √(Σ(x − x̄)² / (n − 1)). The sample version divides by n − 1, not n. The coefficient of variation CV = (s / x̄) × 100% makes spread unit-free for comparison.
- Standard error
- The standard deviation of a sample statistic. For the sample mean it is σ / √n (or s / √n when σ is unknown) - it shrinks as the sample grows, which is why bigger samples give narrower intervals.
- Central Limit Theorem (CLT)
- For a large sample (rule of thumb n ≥ 30) the sampling distribution of x̄ is approximately normal regardless of the population shape, with mean μ and standard error σ / √n. It is what licenses z- and t-based inference.
- Confidence interval
- A range estimate for a parameter: estimate plus or minus (critical value x standard error). A 95% interval means 95% of such intervals over repeated samples would capture the true parameter - not a 95% probability for this one interval.
- z vs t
- Use z when the population sigma is known (or for a proportion); use the t-distribution with df = n - 1 when sigma is estimated by s. The t has heavier tails, so its critical values are larger - the price of estimating the spread.
- Null and alternative hypotheses
- H0 is the status-quo claim and always contains equality; H1 is the research claim. A test gathers evidence against H0; we either reject H0 or fail to reject it - we never accept or prove H0.
- One- vs two-tailed test
- Two-tailed (H1: not equal) splits alpha into both tails and rejects if |statistic| > critical value at alpha/2. One-tailed (H1: > or <) puts all of alpha in a single tail. The chi-square test is upper-tailed only.
- Type I and Type II errors
- A Type I error rejects a true H0 (its probability is alpha); a Type II error fails to reject a false H0 (probability beta). Power = 1 - beta. Lowering alpha raises beta - the two trade off.
- p-value
- The probability, if H0 were true, of a statistic at least as extreme as the one observed. The universal rule: reject H0 if p-value < alpha; otherwise do not reject.
- Chi-square test of independence
- Tests whether two categorical variables are related. Expected count f_e = (row total × column total) / n; statistic χ² = Σ(f_o − f_e)² / f_e on df = (r − 1)(c − 1); it is upper-tailed and needs every f_e ≥ 5.
- Least-squares regression
- Fits the line Ŷ = b₀ + b₁X that minimises the sum of squared vertical residuals. The slope b₁ = Σ(x − x̄)(y − ȳ) / Σ(x − x̄)², and b₀ = ȳ − b₁x̄.
- Coefficient of determination (R²)
- R² = SSR / SST = 1 − SSE / SST is the proportion of variation in Y explained by the model, between 0 and 1. In simple linear regression R² equals the square of the Pearson correlation r.
- MAD and MSFE
- Forecast-accuracy measures for time-series models: MAD = mean of |actual − forecast| and MSFE = mean of (actual − forecast)² . Lower is better; MSFE penalises large errors more heavily because the error is squared.
ETX5900 FAQ
Can AI help me study ETX5900?
Yes. Sia works through Business Statistics with you step by step: paste a workshop question or a past-style problem and Sia explains how to set it up - which test to choose (z vs t, one- vs two-tailed, chi-square, regression), how to state the hypotheses, how to read the standard-normal, t or chi-square table, and how to word the reject or fail-to-reject decision in context.
It explains the reasoning at each step so you learn the method; it will not sit an assessment for you or hand over answers to work you must submit yourself.
Where can I find past exam papers or practice for ETX5900?
Official past papers and the mock e-exam are released inside your Monash Moodle unit for ETX5900, and the Monash Library exam paper collection is the place to check for prior papers - always confirm what is available on Moodle.
To practise the same skills right now, this guide includes a full mini practice exam with fully worked solutions across every topic, and you can ask Sia to generate fresh practice questions in the Part A / B / C style and mark your working step by step.
What can Sia do that a textbook can't?
A textbook shows one worked example; Sia is interactive. It adapts to your exact question, re-explains a step a different way when you are stuck, generates unlimited fresh practice at the difficulty you need, and pinpoints why a specific slip (using z when sigma is unknown, halving alpha on a one-tailed test, the wrong df) cost the mark.
It explains every step of the reasoning so you understand the method - it is a study aid, not a way to get assessment answers or a guaranteed grade.
Is ETX5900 hard?
It is a postgraduate business-statistics unit, so it moves quickly, but it is very learnable because it is procedural: almost every question is the same five-step framework (hypotheses, statistic, critical value, decision, conclusion) with a different distribution.
Because a formula sheet and statistical tables are provided in the exam, the difficulty is in choosing the right procedure and interpreting the result, not in memorising algebra. Steady weekly practice on method selection is what makes it feel straightforward.
Is the ETX5900 exam open or closed book, and is there a hurdle?
The unit materials state that a formula sheet and statistical tables are provided in the final e-exam, but they do not state the open- vs closed-book status or the exam duration (listed as 'to be advised'), and the schedule notes a hurdle requirement as 'See Handbook' without a number.
Do not assume any of these - confirm the book status, duration and any hurdle on Moodle and in the Monash Handbook / unit guide for your teaching period.
What is examined in ETX5900, and how much is the final worth?
The final examination is worth 50% of the unit - an individual, invigilated e-exam in the ~November 2026 end-of-semester period (confirm the date on Moodle). It draws across the whole unit: descriptive statistics, probability, the normal and sampling distributions, confidence intervals, the chi-square test of independence, hypothesis testing, correlation and regression, and time-series forecasting.
The other 50% is within-semester work: in-class activities (15%), take-home quizzes (15%) and a written assignment (20%) in which the chi-square test is a core task.
Questions students ask in each ETX5900 chapter
- How is Descriptive Statistics I examined in ETX5900?
- It underpins the early, short items of the 50% final exam — an individual invigilated e-exam in the roughly November 2026 end-of-semester period (the duration is to be advised, so confirm it on Moodle). A formula sheet and statistical tables are provided, so this material is judgement-tested, not memory-tested: expect to classify variables by type and level, tell a parameter from a statistic, build a small frequency table, and match the correct chart to the data type. Chapter 01 · Descriptive Statistics I: Concepts, Data Types & Visualisation
- When do I divide by n and when by n−1?
- Divide by n−1 for a sample variance or standard deviation (s², s) — the default, since you almost always work from a sample. Divide by N only when the data is the entire population (σ², σ). Using n for a sample is one of the most common mark-losers on this topic, so read whether the question says 'sample' or 'population'. Chapter 02 · Descriptive Statistics II: Distribution, Summary Measures & Pivot Tables
- What is the difference between mutually exclusive and independent events?
- Mutually exclusive means the two events cannot happen together, so P(A and B) = 0. Independent means one event does not change the probability of the other, so P(A and B) = P(A) P(B). They are different ideas: if A and B each have positive probability and are mutually exclusive, then learning A happened makes B impossible, so they are actually dependent — the opposite of independent. Chapter 03 · Probability: Concepts & Applications
- When do I divide by √n and when do I not?
- Divide by √n (use the standard error σ/√n) whenever the question is about a SAMPLE MEAN or average of n observations. Use σ on its own only when the question is about a single individual value X. Mixing these up is the most common lost mark in this module. Chapter 04 · The Normal Distribution & Sampling Distributions
- When do I use z instead of t for a confidence interval for the mean?
- Use z only when the population standard deviation σ is genuinely given. The moment you must estimate the spread from the sample — you are handed s, not σ — switch to t with df = n−1. The t value is larger than the matching z (heavier tails), which widens the interval to account for the extra uncertainty in estimating σ; as n grows, t converges to z. Chapter 05 · Confidence Interval Estimation: Mean & Proportion
- Why is the chi-square test of independence always upper-tailed, and how many degrees of freedom does it have?
- The statistic χ² = Σ(fₒ−fₑ)²/fₑ is a sum of squared, non-negative terms, so it can never be negative and evidence against independence only ever pushes it larger. That means the whole significance level α sits in the right (upper) tail — you reject H₀ when χ²calc exceeds the critical value, and you never split α into two tails. The degrees of freedom come from the table shape: df = (r−1)(c−1), so a 2×2 table has df = 1 and a 3×2 table has df = 2 — it does not depend on the sample size n. Chapter 06 · Chi-Square Test for Independence
- When do I use the Z-test and when the t-test for a single mean?
- It depends on whether the population standard deviation sigma is known. If sigma is known, use Z = (xbar - mu0)/(sigma/sqrt(n)) and the standard normal table. If sigma is unknown and you estimate it with the sample standard deviation s, use t = (xbar - mu0)/(s/sqrt(n)) with df = n - 1 and the t-table. In real business data sigma is almost never known, so the t-test is the usual choice; the t curve has heavier tails, giving slightly larger critical values to reflect the extra uncertainty. Chapter 07 · Hypothesis Testing for Business Decisions I
- Why does the proportion test use pi0 in the standard error instead of the sample proportion p?
- Because a hypothesis test computes the statistic on the assumption that H0 is true, so it uses the value H0 claims: SE = sqrt(pi0(1 - pi0)/n). A confidence interval makes no such assumption, so its margin of error uses the observed p: sqrt(p(1 - p)/n). Mixing them up is a common lost mark - use pi0 in a test and p in a confidence interval. Chapter 08 · Hypothesis Testing for Business Decisions II
- What is the difference between correlation and regression?
- Correlation summarises the relationship in a single number, the Pearson coefficient r, which measures how strong and in which direction two numeric variables move together on a straight line (r between -1 and +1). Regression goes further and fits an actual equation, Y-hat = b0 + b1 X, so you can interpret the slope, predict Y from X, and quantify the fit with R^2. In simple linear regression the two are linked: the sign of r equals the sign of the slope, and R^2 = r^2. Chapter 09 · Introduction to Correlation & Regression Analysis
- Why does the slope t-test use df = n − 2 and not n − 1?
- Because simple linear regression estimates two parameters from the data — the intercept b₀ and the slope b₁ — so two degrees of freedom are used up, leaving n − 2 for the residual variance MSE = SSE/(n−2) and for reading the t-table. (A one-sample mean test estimates a single parameter, so it uses n − 1.) Using n − 1 here reads the wrong row of the t-table and can flip a borderline reject / fail-to-reject decision. Chapter 10 · Regression Model Analysis
- How do I know whether to use a linear or a quadratic trend?
- Plot the series against the time index t. If it rises or falls by a roughly constant amount each period, the linear trend Ŷ_t = b₀ + b₁t fits. If it curves — growth that speeds up or slows down — add the b₂t² term and use the quadratic trend. If you fit both, compare them on MAD and MSFE and keep the model with the lower values, remembering the two measures can occasionally disagree. Chapter 11 · Time Series Analysis & Forecasting
- Is the ETX5900 final exam open-book, and how long is it?
- The final assessment is an individual, invigilated e-exam worth 50% of the unit, sat in the ~November 2026 end-of-semester examination period. A formula sheet and the standard-normal, Student-t and chi-square tables are provided. The unit materials list the duration as 'To be advised' and do not state the open-/closed-book status, so confirm both the exact date and the exam conditions on Moodle / my.Monash before the day. Chapter 12 · Past Exams & Revision
How to study for the exam
Treat ETX5900 as a method-selection unit, not a memorisation unit - the exam provides the formula sheet and statistical tables. Week to week, keep the best-9-of-11 in-class activities and best-10-of-11 quizzes ticking over so the continuous 30% is banked, and start the 20% written assignment (with its core chi-square task on the supplied Excel dataset) early rather than in the final week.
For the 50% final, drill the five-step framework until it is automatic: state H0 and H1, choose the distribution (z if sigma is known, t with df = n - 1 if not; z for a proportion using pi0; chi-square for two categorical variables; the slope t-test with df = n - 2 for regression), get the tail right, read the critical value, and finish with an explicit reject or fail-to-reject decision in context.
Rehearse reading the standard-normal, t and chi-square tables, and do the mock e-exam under conditions to learn the Part A / B / C rhythm. Use SWOTVAC to practise picking the right test on mixed problems, and because the exam length is 'to be advised', practise a mark-proportional pace - spend time on each question in proportion to its marks, and confirm the duration on Moodle.
Your AI Statistics tutor for ETX5900
Stuck on a hard ETX5900 question? Sia is AskSia’s AI Statistics tutor — ask any ETX5900 Business Statistics question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
