Business Forecasting of Double‐trend Time Series: An Improved PLS‐based Time‐varying Weight Combination Approach. (16th November 2017)

• Record Type: Journal Article
• Authors: Luo, Biao; Wan, Liang; Li, Tieshan; Liang, Liang
• Is Part Of: Revue canadienne des sciences de l'administration. Volume 35:Number 3(2018)
• Page Start: 333
• Page End: 348
• Publication Date: 2017-11-16
• Subjects: business forecasting -- partial least squares -- time‐varying weight -- double‐trend time series -- single forecasting -- combination forecasting; prévision conjoncturelle -- moindres carrés partiels -- poids continu temporalisé -- séries chronologiques à double tendance -- prévision simple -- prévision mixte; Management -- Periodicals; Management -- Canada -- Periodicals; 658.005
• DOI: 10.1002/cjas.1465
• Languages: English
• ISSNs: 0825-0383

Abstract

Abstract: Business forecasting with double‐trend time series (long‐term trends and seasonal volatility) has been challenging due to its complexity. Neither a single time series model nor a fixed‐weight combination approach can fully capture the comprehensive information. We address this issue by proposing an improved partial least squares (PLS) based time‐varying weight combination approach. The proposed method can handle the relations both between the single models involved and between single models and time ordering with time‐varying weights. The test on 20 simulated datasets demonstrates the better and more robust performance of the method. We also apply it to three real datasets. The results show that our approach represents a significant improvement over the existing methods in terms of data fitness and prediction accuracy. Copyright © 2017 ASAC. Published by John Wiley & Sons, Ltd.