search query: @author Janacek, G. J. / total: 1
reference: 1 / 1
« previous | next »
Author: | Swift, A. L. Janacek, G. J. |
Title: | Forecasting non-normal time series |
Journal: | Journal of Forecasting
1991 : SEP, VOL. 10:5, p. 501-520 |
Index terms: | TIME SERIES FORECASTING TECHNIQUES STATISTICAL METHODS SIMULATION |
Language: | eng |
Abstract: | The problem of forecasting time series is considered which are not normally distributed. Such data arise as the correlated inter-arrival times of queuing systems, river flow, wind velocities or speech waves etc. An overall approach is suggested working both on simulated data and on real data sets. The approach is based on the autoregressive moving average (ARMA) method, and the models are estimated via a likelihood procedure taking into account the non-normality of the data. The circumstances are examined in which taking explicit account of the non-normality improves the forecasting process in a significant way. Results from several simulated and real series are discussed. |
« previous | next »
SCIMA