Exponential smoothing for time series with more than one cycle?
In exponential smoothing functions (e.g. TESMTH), it will be great if NumXL support double and triple seasonality, just like it does with one seasonality? For example, a power demand time series has daily, weekly and annual cycle. See (4.4) in the attached document.
TripleSeasonalEJOR.pdf
Comments
Please sign in to leave a comment.