If I use Triple Exponential Smoothing with Additive Seasonality and let a statistical program optimize alpha, beta and gamma for me, is there something I can conclude about my data based on the resulting parameters?

E.g. my alpha tends to be close to 0 and my beta close to one for several data sets. (I don't see anything tendency in gamma)

When looking at the formula for level L for example, L is

Lt = α × (Xt − St−L) + (1 − α) × (Lt−1 + Tt−1)

And when alpha is close to zero, the first addend more or less gets omitted. Same with a beta close to one, the second addend gets or less omitted in:

Tt = β × (Lt − Lt−1) + (1 − β) × Tt−1

Could I say a high beta tends to ignore the previous trend, making the trend depend more on the Level, and similar with a low alpha that level does not depend on seasonality? Or am I overinterpreting?

I am relatively new to time series forecasting.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.