This July 2026 ANL252 end-of-course assessment is built around a payment delay dataset. You clean the data in Python, present three charts with insights, model Days_Overdue_Delay with linear regression, evaluate the model and its equation, and suggest further analytics techniques, all within strict word limits per question.

Section A (100 marks) – Answer all questions

The dataset used in this paper contains information about payment delay, and its data dictionary is provided in the Appendix. Please refer to Canvas for details of this dataset.

Notes on assignment writing: Your writing should be succinct but not at the expense of excluding relevant details. The topics in the main report should be presented in the order of the tasks/questions listed in the assignment (Question 1, Question 2, and so on). To avoid a high Turnitin score, do not copy the assignment questions into the report. Some questions may not have absolutely right or wrong answers; for such questions you may express your views about the problem, and you may engage in independent research to demonstrate higher-order thinking skills.

Python code in text format in a monospaced font (for example, Courier New or Lucida Sans Typewriter) must be provided in the main report (the main report excludes the appendices). Less relevant details may be placed in the Appendix. The word limit for each question includes in-text citations and the corresponding content in the appendix (if any), but excludes Python code and the reference list.

Question 1 (30 marks)

Propose at least three (3) data pre-processing methods and apply them using Python to ensure the given payment delay dataset is properly cleaned and prepared. Provide explanations for each task. [Word limit: 300 words]

Question 2 (30 marks)

Use Python to generate three (3) figures based on the processed payment delay dataset from Question 1. Discuss the key insights derived from each figure. Each figure, together with its corresponding Python code and discussion of insights, is worth 10 marks. The figures and their corresponding Python code must be included in the main body of the report as part of the answer. [Word limit: 450 words]

Question 3 (15 marks)

Use linear regression with Python to model the processed dataset obtained from Question 1, with ‘Days_Overdue_Delay’ as the target variable. Explain the modelling approach taken and the decisions made in developing the model. [Word limit: 300 words]

Question 4 (15 marks)

Using Python, evaluate the performance of the linear regression model developed in Question 3, state the resulting regression equation, and discuss the relevant insights. [Word limit: 300 words]

Question 5 (10 marks)

Discuss additional data analytics techniques or modelling approaches that could enhance the insights obtained from the linear regression model above. You may make reasonable assumptions to support your discussion. [Word limit: 300 words]

Appendix: Data Dictionary (ECAPaymentDelay dataset)

Submission Requirements

• Submit one report in Microsoft Word (.docx, Word 2016 or higher) to Canvas via Turnitin by 12 noon on Friday, 30 October 2026.
• Include the course code, ECA title, SUSS PI number, name, submission date and T/TG group on the cover page.
• Python code must be shown in monospaced text for each part of the question; charts are inserted as images in the Word document.
• Generative AI tools are allowed with full attribution (tool name, prompts used, full output and which parts were adapted); SUSS may use a viva voce to confirm authorship.
• Late submissions are subject to the SUSS mark-deduction scheme; references follow APA.