11 Dec Under what conditions would you prefer a simple exponential smoothing model to the moving averages method for forecasting a time series?
1. Under what conditions would you prefer a simple exponential smoothing model to the moving averages method for forecasting a time series?
2. Suppose that monthly data on some time series variable exhibit obvious seasonality. Can you use moving averages, with any appropriate span, to track the seasonality well? Explain why or why not.
3. Suppose that quarterly data on some time series variable exhibit obvious seasonality, although the seasonal pattern varies somewhat from year to year. Which method will work best: Winters’ method or regression with dummy variables for quarters (and possibly a time variable for trend)? Why?
Reference:
Albright, C. S., & Winston, W. L. (2019). Business Analytics: Data Analysis & Decision Making (MindTap Course List) (7th ed.). Cengage Learning
