EUVIMEDEUVIMED entdecken
Crossref

A Comparative Forecasting Framework for Turkey–Germany Trade: Evidence From Time Series and Artificial Neural Networks Models

Seyma Nur Unal, Huseyin Karamelikli

Journal of Forecasting · 2026-04 · 45 (3) · 1177-1187

DOI: 10.1002/for.70084 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Forecasting, online erschienen am 16.12.2025. Verfasst von Seyma Nur Unal und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 43 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper examines the trade relationship between Turkey and Germany by generating strategically consistent forecasts for import and export flows using time series models and artificial neural networks (ANNs). Utilizing monthly trade data from 2002 to 2021, the study compares traditional time series approaches with nonlinear autoregressive exogenous (NARX) ANNs. The results show that a single‐hidden‐layer NARX model with four lags provides the most accurate forecasts across commodity sections, outperforming alternative specifications. Although the models effectively capture overall trade dynamics, the analysis indicates that forecast performance varies across disaggregated sectors. The study demonstrates the usefulness of ANN‐based forecasting for short‐term trade planning while also noting limitations related to data length and model generalizability. The findings offer policy‐relevant insights for improving trade strategy, enhancing early‐warning mechanisms, and supporting data‐driven decision‐making in bilateral trade management.

Zitieren

APA

Unal, S. N., & Karamelikli, H. (2026). A Comparative Forecasting Framework for Turkey–Germany Trade: Evidence From Time Series and Artificial Neural Networks Models. Journal of Forecasting, 45(3), 1177-1187. https://doi.org/10.1002/for.70084

Vancouver

Unal SN, Karamelikli H. A Comparative Forecasting Framework for Turkey–Germany Trade: Evidence From Time Series and Artificial Neural Networks Models. Journal of Forecasting. 2026;45(3):1177-1187. doi:10.1002/for.70084

BibTeX
@article{Unal2026,
  author = {Unal, Seyma Nur and Karamelikli, Huseyin},
  title = {A Comparative Forecasting Framework for Turkey–Germany Trade: Evidence From Time Series and Artificial Neural Networks Models},
  journal = {Journal of Forecasting},
  year = {2026},
  volume = {45},
  number = {3},
  pages = {1177--1187},
  doi = {10.1002/for.70084}
}
Kostenlos starten →