DATA4400 Fitting and evaluating time series models Assignment

Assessment 2 Information

Your Tasks

1. Part A: Submit an Individual report on Forecasting Techniques and Time Series

1. Part B: Quiz (30 minutes): Answer multiple choice questions based on Workshops 1 to 8 material.

Assessment Instructions

Part A Report: Forecasting for Decision-Making: A Data-Driven Approach

Scenario:

You are a Data Scientist at a global analytics firm and have been assigned to forecast trends in a unique timeseries of monthly values (each student will receive different data). Your forecasts will help business leaders optimise decision-making in areas such as demand planning, climate predictions, financial projections, or consumer trends.

Your task is to apply a series of forecasting techniques and produce a report explaining your methodology, results, and recommendations.

You will be provided with a dataset on Monday of Week 7.

Your report should be well-structured, professionally formatted, and include clear visualisations to communicate findings effectively.

You can use software of your choice such as Excel, Exploratory, Orange, Python, or Tableau Public to assist.

Important: If using Tableau Public, you must include screenshots of all your parameters, model evaluation metrics, and settings changes as you will not be able to download the project from Tableau Public to submit the file via the portal.

1. Exponential Smoothing & Holt-Winters Method (3 marks)

• Apply Simple Exponential Smoothing and Holt-Winters (Triple Exponential Smoothing) and forecast for one year ahead. Calculate the MASE for Holt-Winters over the final year, where you define a monthly seasonal naive forecast as the value in the latest corresponding month.
• Interpret all of the analysis above.

2. Prophet and correlogram (5 marks)

• Fit Prophet to the time series with additive seasonals and with multiplicative seasonals. Calculate the RMSE. Continue with the option that gives the lower RMSE.
• Identify seasonality patterns and trends and discuss their impact on forecasting decisions.
• Obtain the Remainder series from your Prophet fit. Calculate the ACF and PACF of the Remainder series. Explain the role of ACF and PACF plots in time series forecasting.
• Interpret all of the analysis above.

3. Seasonal ARIMA Modelling (4 marks)

• Implement seasonal ARIMA models.
• Discuss the model selection criteria you use to choose a suitable model. State the order of your chosen model together with its RMSE
• Forecast for 1 year ahead and provide 68% limits of prediction.
• Compare your chosen seasonal ARIMA results to Holt-Winters and state which model you would recommend in this context.

4. Var models (6 marks)

• What is a VAR model, and when is it used in forecasting?
• Find a real-world application of VAR and discuss how VAR can be implemented.
• Explain the concept of Granger causality and its application in time series forecasting.
• Is there evidence of Granger causality in the VAR application you found?

5. Professionalism (2 Marks)

• Your report should be clear, well-structured, and formatted professionally to effectively communicate your forecasting insights. Include graphs, tables, and statistical outputs to support your discussion. Ensure clear labelling and professional formatting.
• Your referencing must be correct and without the use of Generative AI in the report.
• Your report must be submitted in Word format to the correct portal on time.

REPORT SUBMISSION GUIDELINES

Technical Tools Allowed:

• Excel, Tableau Public, Exploratory, Orange, Python (or any relevant software). If using Tableau Public, you must include screenshots of all your parameters, model evaluation metrics, and

settings changes as you will not be able to download the project from Tableau Public to submit the file via the portal.

File Format:

• Submit as a Word document.
  • Submit other software files.

Part B Quiz:

• The quiz will be available from Monday 10:00 am - Friday 5 pm (AEST) Week 9.
  • You will enter the “Attempt Quiz” on MyKBS in the Assessment Table for A2.
  • Answer 15 multiple-choice questions.
  • Once you start the quiz you will have 30 minutes to complete it.
  • You can start the quiz anytime between 10 am AEST Monday to 5 pm AEST Friday, Week 9. If you miss this window, the quiz will not be opened to you again.
  • The quiz can only be attempted once.
  • Backtracking of questions is not allowed. You must complete the question before moving on to the next one. You will not be able to go back to the previous question.
  • This is an open-book quiz however the use of Generative AI is not permitted.

Generative AI Traffic Lights

Please see the level of Generative AI that this assessment has been designed to accept:

Marking Rubric

Part A Report: Forecasting for Decision-Making: A Data-Driven Approach - Total Marks: 20

Part B Quiz – 10 marks

Note 1: A penalty will be awarded if any part of the assessment is deemed by the assessor where it shows an over-reliance on AI-generated content in your answer. There needs to be a demonstration of original thought.

If you want to challenge the penalty awarded based on Note 1, your assessment will be submitted to Academic Integrity for a second opinion and further investigations.

Important Study Information

Academic Integrity Policy

KBS values academic integrity. All students must understand the meaning and consequences of cheating, plagiarism and other academic offences under the Academic Integrity and Conduct Policy.

What is academic integrity and misconduct? What are the penalties for academic misconduct? What are the late penalties?

How can I appeal my grade?

Click here for answers to these questions: http://www.kbs.edu.au/current-students/student-policies/.

Word Limits for Written Assessments

Submissions that exceed the word limit by more than 10% will cease to be marked from the point at which that limit is exceeded.

Study Assistance

Students may seek study assistance from their local Academic Learning Advisor or refer to the resources on the MyKBS Academic Success Centre page. Click here for this information.

Late assignment submission penalties

Penalties will be imposed on late assignment submissions in accordance with Kaplan Business School’s Assessment Policy.

*Assignments submitted at any stage within the first 24 hours after deadline will be considered to be one day late and therefore subject to the associated penalty.

If you are unable to complete this assessment by the due date/time, please refer to the Special Consideration Application Form, which is available at the end of the KBS Assessment Policy:

https://www.kbs.edu.au/wp- content/uploads/2016/07/KBS_FORM_AssessmentPolicy_MAR2018_FA.pdf