Premium
Stationary Forecasting; Using Holt‐Winter and a Modification of Holt‐Winter
Author(s) -
Kazempour Mohammed K.
Publication year - 1990
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710320317
Subject(s) - mathematics , autoregressive–moving average model , autoregressive integrated moving average , mean squared error , statistics , autoregressive model , moving average , constant (computer programming) , exponential smoothing , mean squared prediction error , smoothing , projection (relational algebra) , mean square , econometrics , time series , algorithm , computer science , programming language
The mean square error of the best (projection) linear predictor and some approximation methods are compared. Some analytical results concerning the optimal ‘smoothing constant’ for Holt‐Winter in Moving Average (1), model MA(1), and Autoregressive Moving Average (1,1), model ARMA(1,1), have been derived. A modified Holt‐Winter technique has been proposed and it is shown for MA(1) and ARMA(1,1) it coincides with the best linear predictor. Finally, we considered flow prediction for the Niger River and the performance of modified Holt‐Winter graphically.