l****a
发帖数: 336 | 1 I am trying to fit data through Holt-Winters(HW) Method with Multiplicative Seasonality
. However, JMP, the software i am allow to use, includes HW Method with Additive
version only.
I do see some literature mentions that for ARIMA process, using Additive
model after logging transforamtion of time seires data is equivalent to the
use of Multiplicative model.
Does the above rule apply to HW Mothed?
Thanks for any tips and help. |
a*****r
发帖数: 681 | 2 proc forecast data=AAA interval=month lead=18 method=winters
seasons=12 out=BBB outfull
outest=est outfitstats;
by org dst;
id yymm;
var ccc;
run; |
a*****r
发帖数: 681 | 3 there is no way to use proc arima to deal with HW model from my point of
view. but SAS do provide proc forecast to "call" HW model. |
l****a
发帖数: 336 | 4 Thanks Aviator.
But I am only allowed to use JMP, which has additive version of winter's
method. :( |
l****a
发帖数: 336 | 5 I will save your sas sample code anyway for future reference.
Thanks again. :) |
