BUSA3015 Business Forecasting
Task:
Exercise 1
For the Seasonally-adjusted data available in Table 1: Forecast food retailing for every month of 2020 using Holt’s Exponential Smoothing with the following parameters: alpha = 0.1 and beta = 0.1. For the seed of the level use the first actual observation. For the seed of the level and the trend – utilise the methods described in class.
Once you perform Holt’s Exponential Smoothing with both alpha and beta equal to 0.1, what are the following numerical values:
1. The within-sample forecast for December 2019.
2. The out-of-sample forecast for January 2020.
3. The out-of-sample forecast for December 2020.
4. The MSE.
5. The MAE.
Use the Solver tool in Excel to calculate the value of alpha and beta that minimises the MSE. After this optimisation, what are the following numerical values:
6. Alpha
7. Beta
8. The MSE
9. The out-of-sample forecast for January 2020.
10. The out-of-sample forecast for December 2020.
Exercise 2
For the Original data available in Table 1: Forecast food retailing for every month of 2020 using Winter’s Exponential Smoothing (Multiplicative) with the following parameters: alpha = 0.1, beta = 0.1, and gamma = 0.1. For the seeds of the level, trend, and seasonal components – utilise the methods described in class.
Once you perform Winter’s Exponential Smoothing with alpha, beta, and gamma all equal to 0.1, what are the following numerical values:
11. The seasonal component for December 2019. 12. The within-sample forecast for December 2019.
13. The out-of-sample forecast for December 2020.
14. The MSE.
15. The MAE.
Use the Solver tool in Excel to calculate the value of alpha, beta, and gamma that minimises the MSE.
After this optimisation, what are the following numerical values:
16. Alpha
17. Gamma
18. The MSE
19. The within-sample forecast for December 2019.
20. The out-of-sample forecast for December 2021.
Exercise 3
For Exercise 1 – Label this “Critical Analysis – Exercise 1”
21. Plot a chart that contains both the sample data (January 2010 – December 2019) and the forecasts (January 2020 – December 2020) – where there is a clear distinction between the sample data and the forecasts.
22. Critically analyse and comment on your smoothing.
23. Perform the appropriate check/s and test/s that help critically analyse whether your model has captured all the systematic components and/or whether the errors are random. Explain your answer.
For both Exercise 1 and Exercise 2 – Label this “Critical Analysis, Evaluation, and Recommendation”
27. Compare and contrast the two models and critically analyse your results and put forward a recommendation for your choice of model.
28. Do any of your results (for any of the above questions) suggest any re-evaluation or modification of the forecasting method/s, if at all? Explain your answer.
29. In light of recent events, critically evaluate your choice of model, and critically evaluate the factors you would need to consider when forecasting in light of recent events.
30. Given that you have some of the actual data available for 2020, critically evaluate your forecasts and forecasting method.
31. In the context of business forecasting, critically think and discuss any other considerations that need to be taken into account for your forecasts / forecasting to be useful for business purposes.
