PCA-BASED EX-ANTE FORECASTING OF SWAP TERM STRUCTURES

Oliver Blaskowitz and Helmut Herwartz ()
Additional contact information
Helmut Herwartz: Institute of Statistics and Econometrics, Christian-Albrechts-Universität zu Kiel, 24098 Kiel, Germany

International Journal of Theoretical and Applied Finance (IJTAF), 2009, vol. 12, issue 04, pages 465-489

Abstract: In this study, we forecast the term structure of EURIBOR swap rates by means of rolling vector autoregressive (VAR) models. In advance, a principal component analysis (PCA) is adopted to reduce the dimensionality of the term structure. To statistically assess the forecasting performance for particular rates and the level, slope and curvature of the swap term structure, we rely on the Henrikkson–Merton statistic. Economic performance is investigated by means of cash flows implied by alternative trading strategies. Finally, a data-driven, adaptive model selection strategy to "predict the best forecasting model" out of a set of 100 alternative PCA/VAR implementations is shown to outperform forecasting schemes that rely on global homogeneity of the term structure.

Keywords: Adaptive ex-ante forecasting; EURIBOR swap rates; term structure (search for similar items in EconPapers)
Date: 2009

Downloads: (external link)
http://www.worldscinet.com/cgi-bin/details.cgi?type=pdf&id=pii:S021902490900535X (application/pdf)
http://www.worldscinet.com/cgi-bin/details.cgi?typ ... ii:S021902490900535X (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:wsi:ijtafx:v:12:y:2009:i:04:p:465-489

Ordering information: This journal article can be ordered from

Access Statistics for this article

International Journal of Theoretical and Applied Finance (IJTAF) is edited by L P Hughston

More articles in International Journal of Theoretical and Applied Finance (IJTAF) from World Scientific Publishing Co. Pte. Ltd.
Series data maintained by Tai Tone Lim ().