Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Forecasting the yield curve of government bonds: a dynamic factor approach

Forecasting the yield curve of government bonds: a dynamic factor approach <jats:sec> <jats:title content-type="abstract-subheading">Purpose</jats:title> <jats:p>Forecasting the future movement of yield curves contains valuable information for both academic and practical issues such as bonding pricing, portfolio management, and government policies. The purpose of this paper is to develop a dynamic factor approach that can provide more precise and consistent forecasting results under various yield curve dynamics.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Design/methodology/approach</jats:title> <jats:p>The paper develops a unified dynamic factor model based on Diebold and Li (2006) and Nelson and Siegel (1987) three-factor model to forecast the future movement yield curves. The authors apply the state-space model and the Kalman filter to estimate parameters and extract factors from the US yield curve data.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Findings</jats:title> <jats:p>The authors compare both in-sample and out-of-sample performance of the dynamic approach with various existing models in the literature, and find that the dynamic factor model produces the best in-sample fit, and it dominates existing models in medium- and long-horizon yield curve forecasting performance.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Research limitations/implications</jats:title> <jats:p>The authors find that the dynamic factor model and the Kalman filter technique should be used with caution when forecasting short maturity yields on a short time horizon, in which the Kalman filter is prone to trade off out-of-sample robustness to maintain its in-sample efficiency.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Practical implications</jats:title> <jats:p>Bond analysts and portfolio managers can use the dynamic approach to do a more accurate forecast of yield curve movements.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Social implications</jats:title> <jats:p>The enhanced forecasting approach also equips the government with a valuable tool in setting macroeconomic policies.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Originality/value</jats:title> <jats:p>The dynamic factor approach is original in capturing the level, slope, and curvature of yield curves in that the decay rate is set as a free parameter to be estimated from yield curve data, instead of setting it to be a fixed rate as in the existing literature. The difference range of estimated decay rate provides richer yield curve dynamics and is the key to stronger forecasting performance.</jats:p> </jats:sec> http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Managerial Finance CrossRef

Forecasting the yield curve of government bonds: a dynamic factor approach

Managerial Finance , Volume 43 (7): 774-793 – Jul 10, 2017

Forecasting the yield curve of government bonds: a dynamic factor approach


Abstract

<jats:sec>
<jats:title content-type="abstract-subheading">Purpose</jats:title>
<jats:p>Forecasting the future movement of yield curves contains valuable information for both academic and practical issues such as bonding pricing, portfolio management, and government policies. The purpose of this paper is to develop a dynamic factor approach that can provide more precise and consistent forecasting results under various yield curve dynamics.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Design/methodology/approach</jats:title>
<jats:p>The paper develops a unified dynamic factor model based on Diebold and Li (2006) and Nelson and Siegel (1987) three-factor model to forecast the future movement yield curves. The authors apply the state-space model and the Kalman filter to estimate parameters and extract factors from the US yield curve data.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Findings</jats:title>
<jats:p>The authors compare both in-sample and out-of-sample performance of the dynamic approach with various existing models in the literature, and find that the dynamic factor model produces the best in-sample fit, and it dominates existing models in medium- and long-horizon yield curve forecasting performance.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Research limitations/implications</jats:title>
<jats:p>The authors find that the dynamic factor model and the Kalman filter technique should be used with caution when forecasting short maturity yields on a short time horizon, in which the Kalman filter is prone to trade off out-of-sample robustness to maintain its in-sample efficiency.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Practical implications</jats:title>
<jats:p>Bond analysts and portfolio managers can use the dynamic approach to do a more accurate forecast of yield curve movements.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Social implications</jats:title>
<jats:p>The enhanced forecasting approach also equips the government with a valuable tool in setting macroeconomic policies.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Originality/value</jats:title>
<jats:p>The dynamic factor approach is original in capturing the level, slope, and curvature of yield curves in that the decay rate is set as a free parameter to be estimated from yield curve data, instead of setting it to be a fixed rate as in the existing literature. The difference range of estimated decay rate provides richer yield curve dynamics and is the key to stronger forecasting performance.</jats:p>
</jats:sec>

Loading next page...
 
/lp/crossref/forecasting-the-yield-curve-of-government-bonds-a-dynamic-factor-1LbV0tfwPJ
Publisher
CrossRef
ISSN
0307-4358
DOI
10.1108/mf-11-2016-0330
Publisher site
See Article on Publisher Site

Abstract

<jats:sec> <jats:title content-type="abstract-subheading">Purpose</jats:title> <jats:p>Forecasting the future movement of yield curves contains valuable information for both academic and practical issues such as bonding pricing, portfolio management, and government policies. The purpose of this paper is to develop a dynamic factor approach that can provide more precise and consistent forecasting results under various yield curve dynamics.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Design/methodology/approach</jats:title> <jats:p>The paper develops a unified dynamic factor model based on Diebold and Li (2006) and Nelson and Siegel (1987) three-factor model to forecast the future movement yield curves. The authors apply the state-space model and the Kalman filter to estimate parameters and extract factors from the US yield curve data.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Findings</jats:title> <jats:p>The authors compare both in-sample and out-of-sample performance of the dynamic approach with various existing models in the literature, and find that the dynamic factor model produces the best in-sample fit, and it dominates existing models in medium- and long-horizon yield curve forecasting performance.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Research limitations/implications</jats:title> <jats:p>The authors find that the dynamic factor model and the Kalman filter technique should be used with caution when forecasting short maturity yields on a short time horizon, in which the Kalman filter is prone to trade off out-of-sample robustness to maintain its in-sample efficiency.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Practical implications</jats:title> <jats:p>Bond analysts and portfolio managers can use the dynamic approach to do a more accurate forecast of yield curve movements.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Social implications</jats:title> <jats:p>The enhanced forecasting approach also equips the government with a valuable tool in setting macroeconomic policies.</jats:p> </jats:sec> <jats:sec> <jats:title content-type="abstract-subheading">Originality/value</jats:title> <jats:p>The dynamic factor approach is original in capturing the level, slope, and curvature of yield curves in that the decay rate is set as a free parameter to be estimated from yield curve data, instead of setting it to be a fixed rate as in the existing literature. The difference range of estimated decay rate provides richer yield curve dynamics and is the key to stronger forecasting performance.</jats:p> </jats:sec>

Journal

Managerial FinanceCrossRef

Published: Jul 10, 2017

References