Module 4 – Case INSTRUCTIONS
RISK: Assignment Overview
Scenario: You are still a consultant for the Excellent Consulting Group. You have completed the first assignment, developing and testing a forecasting method based on linear regression (Case 3). However, your consulting manager at ECG wants to go the next step and investigate another forecasting method. It is important to do a thorough job for the client, and you have the expertise to analyze different forecasting methods. You have decided to look at the sales data for client’s lottery app as a single data set and use a time series analysis, namely SES, single exponential smoothing.
Case Assignment
Using Excel, use the forecasted sales from Case 3 to compute the MAPE, by doing the following:
1. Calculate the MAPE for the first 12 months (assume the forecast for Month 1 – or January – is equal to January’s actual sales). Use 0.15 and 0.90 alphas.
2. Using the forecasted sales for Feb – April (taken from the Case 3 Linear Regression exercise), compute the MAPE by comparing actual sales for each month, or Y(t) to forecasted sales, or F(t). Compare this 3-month MAPE to the two MAPE values you calculated in your SES analysis above. Use the following table:
Month Sales, Y(t) Sales F(t) Y(t) – F(t) PE APE
February ? ? ? ? ?
March ? ? ? ? ?
April ? ? ? ? ?
? ? ?
ME MPE MAPE
Then write a report to your boss that briefly describes the results that you obtained. Using MAPE values, make a recommendation on which method appears to be more accurate — SES or Linear Regression.
Data: Use the data that you previously have generated from your analyses in Case 3.
Assignment Expectations
Analysis
• Accurate and complete SES analysis in Excel.
Written Report
• Length requirements = 4–5 pages minimum (not including Cover and Reference pages)
• Provide a brief introduction/ background of the problem.
• Complete and accurate Excel analysis.
• Written analysis that supports Excel analysis, and provides thorough discussion of assumptions, rationale, and logic used.
• Complete, meaningful, and accurate recommendation(s).