Assignment

Forecasting Assignment

Submit:
1) a document of your answers, and
2) your completed Forecasting_Data.xlsx

1. Suppose we are forecasting next month’s sales for an item. Which forecast would you expect to have the lowest mean absolute percent error (MAPE), a forecast for one particular store or a forecast across three stores in the region? Explain.

1. Suppose we are forecasting using weekly time intervals. What do you call a forecast made at the end of week 2 for week 5?

1. Suppose a firm is using simple exponential smoothing to forecast for a demand series where there is no seasonality but there is a downward trend. If they don’t adjust for this trend, will the forecasts tend to be too high or too low? Explain.

For questions 4-6, refer to the sheet ForecastPerformance. Label your work in the file.

1. What is the average absolute forecast error over the first 10 weeks?

1. What is the average forecast error over the first 5 weeks?

1. After 10 weeks, is there is evidence to suggest that this forecasting process is biased? Explain.

For questions 7-9, refer to the sheet ExponentialSmoothing. Label your work in the file.

1. What is the lag 2 forecast made at the end of week 1 (i.e. what is the forecast made at end of week 1 for week 3)?

1. What is the forecast for week 4 made at the end of week 3?

1. At the end of week 10, there is evidence to suggest this forecasting process is biased. From the list below, which would be the best option for addressing this bias? Explain why your choice is better than either of the other two options.

• Add a trend component to the forecasting model

• Adjust for an outlier in the data

• Adjust the initial forecasting estimate

For questions 10-11, refer to the sheet Trend in the Forecasting_Data Excel file. Label your work into the Excel file.

1. What is the trend estimate at the end of month 1?

2. What is the forecast for month 4 at the end of month 2?