Seasonal Model and Its Application in Short-term Forecasting

DOI

10.2991/ammsa-18.2018.39How to use a DOI?

Keywords

seasonality; time series; forecasting; ARIMA

Abstract

With economic activity changes periodically as the seasons vary, the time series of macroeconomic data is usually seasonal. This paper is different from the traditional way of proceeding seasonal adjustment before analyzing the macroeconomic time series. Instead, we propose a seasonal prediction model based on the stable seasonal variation pattern of time series to conduct short-term prediction. The basic principle of the model is to first establish the conditional distribution of the total accumulation based on the known observed values and to estimate the seasonal parameters on historical data using the maximum likelihood method. And then conduct prediction with the observations observed and the estimated parameters for the seasonal pattern. The model shows high prediction efficiency for the time series with stable seasonality through the empirical analysis.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xiangrong Jiang
AU  - Liping Xu
AU  - Yingying Cui
PY  - 2018/05
DA  - 2018/05
TI  - Seasonal Model and Its Application in Short-term Forecasting
BT  - Proceedings of the 2018 2nd International Conference on Applied Mathematics, Modelling and Statistics Application (AMMSA 2018)
PB  - Atlantis Press
SP  - 194
EP  - 196
SN  - 1951-6851
UR  - https://doi.org/10.2991/ammsa-18.2018.39
DO  - 10.2991/ammsa-18.2018.39
ID  - Jiang2018/05
ER  -