EUVIMEDEUVIMED entdecken
Crossref

Yield curve forecasting using the Nelson-Siegel model. A comparison of ARIMA, VAR and Random Forest approaches – evidence from the USA

Maciej Olszewski

Przegląd Statystyczny. Statistical Review · 30.06.2026 · 2026 (2) · 16-34

DOI: 10.59139/ps.2026.02.2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Przegląd Statystyczny. Statistical Review, online erschienen am 30.06.2026. Verfasst von Maciej Olszewski.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The aim of this article is to compare different approaches to forecasting the US yield curve factors derived using the Nelson-Siegel (NS) model. Using daily US swap yield data from 1990 to 2026, we assess the Autoregressive Integrated Moving Average (ARIMA), Vector Autoregression (VAR) and Random Forest (RF) models in a 1-day-ahead and 1- to 20-day-ahead forecasting competition. The principal finding of this study is that ARIMA significantly outperforms the RF in forecasting the NS model factors, as does VAR, although only in terms of the Level and Curvature factors. The results of this study thus suggest that the use of machine learning methods, in the case of the US yield curve, is not always superior.

Zitieren

APA

Olszewski, M. (2026). Yield curve forecasting using the Nelson-Siegel model. A comparison of ARIMA, VAR and Random Forest approaches – evidence from the USA. Przegląd Statystyczny. Statistical Review, 2026(2), 16-34. https://doi.org/10.59139/ps.2026.02.2

Vancouver

Olszewski M. Yield curve forecasting using the Nelson-Siegel model. A comparison of ARIMA, VAR and Random Forest approaches – evidence from the USA. Przegląd Statystyczny. Statistical Review. 2026;2026(2):16-34. doi:10.59139/ps.2026.02.2

BibTeX
@article{Olszewski2026,
  author = {Olszewski, Maciej},
  title = {Yield curve forecasting using the Nelson-Siegel model. A comparison of ARIMA, VAR and Random Forest approaches – evidence from the USA},
  journal = {Przegląd Statystyczny. Statistical Review},
  year = {2026},
  volume = {2026},
  number = {2},
  pages = {16--34},
  doi = {10.59139/ps.2026.02.2}
}
Kostenlos starten →