Reference no: EM133743691 , Length: word count:500

Data-driven Decision Making and Forecasting

Assessment - A Forecasting Project

Your Task
Apply forecasting techniques to a given dataset and provide a business application of the forecasts.

Assessment Description

The data provided for the assessment are time series of monthly sales revenue for two stores: TheDon which sells cricket equipment; and WorkOut which sells equipment for a wide range of sports.'
The objective of the assessment is to use Exploratory, Tableau and Excel software to: describe the time series; to develop Prophet and Holt-Winters demand forecast models for the two stores; and to compare forecasts from these models.

Assessment Instructions

In class: You will be presented with a dataset in class. As a group, analyse the dataset using Tableau and Exploratory.io.

You will provide an oral presentation of the group work in parts A to D during the third hour of the workshop.

As a group:

Part A

Use Tableau to compare time series plots for the two stores by: plotting one under the other; and by plotting both on the same graph. Explain how you implement these plots in Tableau.

Comment on features of the two time series. Use descriptive analytics to compare the two stores in terms of sales revenue, and explain how these statistics can be obtained using Tableau.

Part B

Use Tableau to investigate trend and seasonal variability in the time series for the two stores. Explain how to implement relevant graphics in Tableau. Comment on the trends and seasonal effects.

Part C

Use Tableau to obtain Holt-Winters forecasts for TheDon for the next 12 months, and to provide a suitable graphical display. Explain the various summaries provided by Tableau, in the context of TheDon. Explain how to find the point forecast three months ahead and its 95% Prediction Interval in Tableau.

Part D

Use Exploratory to obtain Prophet forecasts for TheDon for the next 12 months, and to provide a suitable graphical display. Explain the various summaries provided by Exploratory, in the context of TheDon. Explain how to find the point forecast three months ahead and its 95% Prediction Interval in Exploratory.

Part E

Prepare a PowerPoint presentation with your answers to A, B, C, and D:
Include screenshots showing how you use the software, and of the information provided by the software.
Discuss the suitability of HW and Prophet forecasting for TheDon.

Submit your slide deck via the Turnitin portal + zip your data files along with the analytics model files and submit them via the File Submission portal at the end of the allocated time. Uploading files to the incorrect place or not uploading the data/model files late/at all will be treated as poor project management practice and penalised accordingly.

As individuals:

Take the most recent year as a test series. Use your individual time series, less the final year, as a training series.
a) Plot training series plus HW forecasts for test year for TheDon in Tableau.
b) Plot training series plus Prophet forecasts for test year for TheDon in Exploratory.
c) Selecting from statistics given in Tableau and Exploratory, construct a table of RMSE within training series by method by store.
d) Make calculations in Excel, and construct a table of RMSE in the test series by method by store.
e) Explain why the RMSE is expected to be larger in the test series. Will the RMSE necessarily be larger in the test series?
f) Make calculations in Excel, and find the MASE using Prophet in the test series for WorkOut.