Forecasting using SAS Forecast Server, SAS/ETS, and more

Linear exponential smoothing with PROC ESM

Reply
New Contributor
Posts: 4

Linear exponential smoothing with PROC ESM

I would like to know how to interpret the parameter estimates in PROC ESM when Holt's exponential smoothing method is used.

 

proc esm data=t print=all ;
forecast x /model=linear;
run;

 

The estimate for Trend Weight is always 0.001.

 

 

trend.png 

i applied to various real data sets and the simulated data sets with different parameter settings.

 

 

data t;
a=0.5;
b=0.1;
do t=0 to 100;
x=a+b*t+rannor(123)*0.1;
output;
end;
run;

 

 how can they be converted to the parameters, alpha and gamma, specified in the document?

 

holt.png

 

 thanks.

 

 

New Contributor
Posts: 4

Re: Linear exponential smoothing with PROC ESM

i answer my own question.

Level weight is alpha and Trend weight is gamma.

it just happens that the all data sets i used have constant trends (T(t) = T(t-1)).

the simulation data, of course, have constant trend.

therefore, gamma=0.001 is essentially zero.

 

Ask a Question
Discussion stats
  • 1 reply
  • 106 views
  • 0 likes
  • 1 in conversation