Business and economic statistics Notes
In this unit, you will learn how to analyse and interpret data to support evidence-based decision-making in business and economics. You will explore methods for summarising data, including tables, charts, and measures of central tendency and dispersion. Key topics include standardisation techniques, regression analysis, and the use of probability and sampling to understand uncertainty. The unit also covers statistical inference, including confidence intervals and hypothesis testing. Throughout, you’ll apply these concepts to real-world scenarios, developing practical skills in data analysis.
Chapter-Wise Module Notes & PDF Downloads
Module 1
Descriptive Analytics, Probability Distributions & Random Variables
Module 2
Sampling Distributions, the Central Limit Theorem & Interval Estimation
Module 3
Hypothesis Testing: One-Sample, Two-Sample & Analysis of Variance (ANOVA)
Module 4
Linear Regression Analysis, OLS Estimation, Correlation & Residual Diagnostics
Read Module Notes Online
Monash University Academic Framework
Syllabus details for Bachelor of Commerce (ETC1000) under official Monash University regulations (2026 Monash Handbook (ETC1000)).
Assessment comprises supervised examination and continuous coursework with a 45% hurdle requirement.
About Business and economic statistics (Monash University ETC1000)
Business and Economic Statistics introduces students to the quantitative principles of descriptive statistics, probability theory, sampling distributions, inferential hypothesis testing, and linear regression analysis within commercial and macroeconomic contexts.
The curriculum begins by formalizing data exploration, contrasting measures of central tendency (mean, median, mode) and dispersion (variance, standard deviation, interquartile range), and evaluating skewness and kurtosis. Students advance into discrete probability distributions (Binomial, Poisson) and continuous distributions (Uniform, Normal). The foundational milestone of the unit is the Central Limit Theorem, which enables inferential deduction regarding unknown population parameters using sample statistics.
ETC1000 Module-by-Module Curriculum Breakdown
Official syllabus unit modules prescribed by the Monash University academic handbook:
Descriptive Analytics, Probability Distributions & Random Variables
Module 1 establishes the mathematical groundwork of empirical data exploration and probability theory. Students master numerical summaries: mean, weighted mean, sample variance ($s^2$), and sample standard deviation ($s$). The curriculum deconstructs Chebyshev's theorem and the Empirical Rule for bell-shaped distributions. Students interrogate the axioms of probability: addition rule, multiplication rule, conditional probability $P(A|B) = P(A \cap B)/P(B)$, and Bayes' theorem. The syllabus formalizes discrete random variables via expected value $E(X) = \sum x P(x)$ and variance, exploring the Binomial distribution $B(n, p)$. The module transitions into continuous probability, centering on the Normal distribution $N(\mu, \sigma^2)$ and standardization via Z-scores $Z = (X - \mu)/\sigma$.
Sampling Distributions, the Central Limit Theorem & Interval Estimation
Module 2 bridges descriptive sample data and population inference. Students analyze the sampling distribution of the sample mean $\bar{X}$, proving that $E(\bar{X}) = \mu$ and standard error $\sigma_{\bar{x}} = \sigma / \sqrt{n}$. The Central Limit Theorem (CLT) is mathematically demonstrated: as sample size $n$ increases ($n \ge 30$), the distribution of $\bar{X}$ approaches normality regardless of population distribution shape. The module constructs point estimators and two-sided confidence intervals for $\mu$. When population variance $\sigma^2$ is unknown, students transition from the standard normal Z-distribution to Student's t-distribution with $n - 1$ degrees of freedom. Students evaluate the trade-off between confidence level ($1 - \alpha$) and interval precision (margin of error).
Hypothesis Testing: One-Sample, Two-Sample & Analysis of Variance (ANOVA)
Module 3 engages the formal Neyman-Pearson statistical decision framework. Students formulate mutually exclusive null ($H_0$) and alternative ($H_1$) hypotheses. The syllabus evaluates the duality between Type I error (rejecting true $H_0$ with probability $\alpha$) and Type II error (failing to reject false $H_0$ with probability $\beta$). Students execute five-step hypothesis tests using both critical value rejection regions and exact p-values: one-sample Z-tests and t-tests for population means; two-sample independent t-tests (equal and unequal variance assumptions); and paired-difference t-tests for repeated measures. The module extends to One-Way Analysis of Variance (ANOVA), partitioning total sum of squares into Between-Group and Within-Group variance to test equality of three or more group means using the F-statistic.
Linear Regression Analysis, OLS Estimation, Correlation & Residual Diagnostics
Module 4 formalizes linear econometric modeling. Students analyze Pearson's correlation coefficient ($r$), establishing that correlation measures linear association rather than causal direction. The module derives Ordinary Least Squares (OLS) simple linear regression $Y_i = \beta_0 + \beta_1 X_i + \epsilon_i$, solving for slope ($b_1 = \text{Cov}(X,Y)/s_x^2$) and intercept ($b_0 = \bar{Y} - b_1 \bar{X}$) by minimizing the sum of squared residuals (SSE). Students decompose total variation ($SST = SSR + SSE$) to calculate the coefficient of determination ($R^2 = SSR / SST$). The syllabus covers t-testing of slope significance ($H_0: \beta_1 = 0$), F-testing of overall model utility, and residual diagnostic verification (testing linearity, homoskedasticity, normality, and absence of autocorrelation).
Core Curriculum & Key Topics
- Central Limit Theorem: For independent identically distributed samples of sufficient size (n >= 30), the sampling distribution of the sample mean approaches normality regardless of underlying population shape.
- Neyman-Pearson Hypothesis Testing Framework: Evaluating null hypotheses (H0) against research alternatives (H1) by controlling Type I error (alpha) and minimizing Type II error (beta).
- Gauss-Markov Theorem: Under classical linear regression assumptions, Ordinary Least Squares (OLS) estimators are Best Linear Unbiased Estimators (BLUE).
- Law of Large Numbers: As sample size expands toward infinity, the empirical sample mean converges in probability to the true population expectation.
Learning Outcomes (ULOs)
- ULO1: analyse business and economic data using quantitative methods to inform evidence-based decision-making
- ULO2: apply probability concepts to model uncertainty and assess risk in business and economic contexts
- ULO3: analyse multi-variable data to draw conclusions for business and economic decision-making
- ULO4: communicate statistical findings using appropriate formats, collaborating in teams to present evidence-based insights for business audiences
- ULO5: evaluate academic and industry sources to justify conclusions in business and economic contexts.
Monash University Examination Scheme for ETC1000
The evaluation framework for Business and economic statistics (ETC1000) combines Continuous Assessment (50% Marks) with the supervised End-of-Semester Examination (50% Marks) for a total of 100 Marks (120 minutes duration):
Short definitional and conceptual recognition questions covering core principles across all 4 syllabus modules.
Multi-part ledger accounting, statistical derivations, or comparative evaluation questions with diagnostic computations.
Long-form problem scenarios, financial statement construction, or comprehensive numerical workouts.
Exam Preparation Strategy:
Always write null and alternative hypotheses clearly ($H_0$ vs $H_1$). State your decision rule (e.g. reject $H_0$ if $p < 0.05$) and give a final conclusion in the context of the question. Ensure you meet the 45% examination hurdle requirement by demonstrating systematic working and clear notation.
Exam Preparation & Study Guides — ETC1000
What is the critical distinction between the Standard Deviation and the Standard Error of the Mean?
The standard deviation (s or sigma) measures the dispersion or variability of individual raw data points around the population or sample mean. Conversely, the standard error of the mean (SE = sigma / sqrt(n)) measures the dispersion or variability of sample means across repeated theoretical samplings of size n from the population. As sample size n increases, the standard deviation of the population remains constant, whereas the standard error of the mean strictly decreases toward zero, reflecting increased precision in estimating the population mean.
Explain why correlation does not imply causation in bivariate linear regression.
A high Pearson correlation coefficient (r) proves only that two variables exhibit a co-movement pattern. It does not establish causal direction because three confounding factors frequently produce spurious correlation: (1) reverse causality (Y causes X rather than X causing Y); (2) omitted variable bias, where an unobserved third variable (Z) drives simultaneous changes in both X and Y; and (3) coincidental time-series trends. Establishing true causality requires controlled experiments, instrumental variables, or randomized natural experiment designs.
How do financial institutions use two-sample hypothesis testing in credit risk evaluation?
Credit risk teams employ two-sample t-tests and difference-in-proportions tests to determine whether loan applicants approved under a new automated algorithmic credit-scoring model exhibit a significantly lower 90-day default rate compared to those evaluated under legacy manual underwriting. By testing $H_0: p_1 - p_2 = 0$ against $H_1: p_1 < p_2$ across large randomized applicant cohorts, the risk committee verifies whether the default reduction is statistically attributable to algorithm superior screening rather than random macroeconomic variation.
What happens to OLS estimator properties when the assumption of homoskedasticity is violated?
When the error variance is non-constant across observations (heteroskedasticity), the Ordinary Least Squares (OLS) slope and intercept estimators remain mathematically unbiased, meaning $E(b_1) = \beta_1$. However, OLS is no longer BLUE (Best Linear Unbiased Estimator) because it ceases to have minimum variance. Crucially, the standard errors calculated by standard software are biased and typically underestimated, which artificially inflates t-statistics and leads researchers to incorrectly reject true null hypotheses. Heteroskedasticity-robust standard errors (White's standard errors) must be deployed.
Course outlines and study notes follow official Monash University handbook specifications (2026 Monash Handbook (ETC1000)). Curated by DegreeLive Econometrics & Business Statistics Editorial Board (Commerce Academic Contributor).
For feedback or corrections, contact DegreeLive Academic Support.
Related Study Resources & Student Support
Related Semester 1 Units
Browse study notes for other courses in Commerce:
Financial Aid & Grants 2026-27
Check eligibility criteria and deadlines for university merit scholarships and bursaries.
Request Notes
Can't find what you're looking for? Let us know which unit notes or past papers you need.
