Triple Exponential Smoothing with a high frequency

Machine Learning and Modeling
, ,
1

Hello everyone,

I am trying to develop a forecasting method that is equal to triple exponential smoothing. I cannot use ets() and hw() because my frequency is pretty high with 52 (weekly basis). Therefore I currently use stlf(). However, I am not quite able to figure out how this is mathematically composed in order to interpret it sensibly.

tes <- stlf(train, 
            s.window="periodic", 
            method = "ets",
            lambda = "auto",
            biasadj = TRUE)

# Output
Forecast method: STL +  ETS(A,Ad,N)

Model Information:
ETS(A,Ad,N) 

Call:
 ets(y = na.interp(x), model = etsmodel, allow.multiplicative.trend = allow.multiplicative.trend) 

  Smoothing parameters:
    alpha = 0.0137 
    beta  = 0.0137 
    phi   = 0.9296 

  Initial states:
    l = 2416295.7805 
    b = -1190.9562 

  sigma:  1904907

Is there a possibility to use the triple exponential smoothing according to Holt-Winter's to simply get the three factors (alpha, beta, gamma) and to simplify the comparison to the single and double exponential smoothing?

Thanks a lot for your help.

Luke

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.