EUVIMEDEUVIMED entdecken
Crossref

Nonstationary Functional Time Series Forecasting

Han Lin Shang, Yang Yang

Journal of Forecasting · 2025-07 · 44 (4) · 1347-1362

DOI: 10.1002/for.3241 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Forecasting, online erschienen am 17.12.2024. Verfasst von Han Lin Shang und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 63 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose a nonstationary functional time series forecasting method with an application to age‐specific mortality rates observed over the years. The method begins by taking the first‐order differencing and estimates its long‐run covariance function. Through eigendecomposition, we obtain a set of estimated functional principal components and their associated scores for the differenced series. These components allow us to reconstruct the original functional data and compute the residuals. To model the temporal patterns in the residuals, we again perform dynamic functional principal component analysis and extract its estimated principal components and the associated scores for the residuals. As a byproduct, we introduce a geometrically decaying weighted approach to assign higher weights to the most recent data than those from the distant past. Using the Swedish age‐specific mortality rates from 1751 to 2022, we demonstrate that the weighted dynamic functional factor model can produce more accurate point and interval forecasts, particularly for male series exhibiting higher volatility.

Zitieren

APA

Shang, H. L., & Yang, Y. (2025). Nonstationary Functional Time Series Forecasting. Journal of Forecasting, 44(4), 1347-1362. https://doi.org/10.1002/for.3241

Vancouver

Shang HL, Yang Y. Nonstationary Functional Time Series Forecasting. Journal of Forecasting. 2025;44(4):1347-1362. doi:10.1002/for.3241

BibTeX
@article{Shang2025,
  author = {Shang, Han Lin and Yang, Yang},
  title = {Nonstationary Functional Time Series Forecasting},
  journal = {Journal of Forecasting},
  year = {2025},
  volume = {44},
  number = {4},
  pages = {1347--1362},
  doi = {10.1002/for.3241}
}
Kostenlos starten →