scholarly journals A Dynamic Factor Model of the Yield Curve as a Predictor of the Economy

2012 ◽  
Vol 2012 (32) ◽  
Author(s):  
Marcelle Chauvet ◽  
◽  
Zeynep Senyuz
Keyword(s):  
Yield Curve ◽  
Factor Model ◽  
Dynamic Factor ◽  
Dynamic Factor Model

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

2017 ◽  
Vol 43 (7) ◽  
pp. 774-793
Author(s):  
Walid Ben Omrane ◽  
Chao He ◽  
Zhongzhi Lawrence He ◽  
Samir Trabelsi
Keyword(s):  
Kalman Filter ◽  
Yield Curve ◽  
Factor Model ◽  
Dynamic Factor ◽  
Dynamic Factor Model ◽  
Dynamic Approach ◽  
Yield Curves ◽  
Content Type ◽  
Forecasting Performance ◽  
Out Of Sample

Purpose 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. Design/methodology/approach 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. Findings 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. Research limitations/implications 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. Practical implications Bond analysts and portfolio managers can use the dynamic approach to do a more accurate forecast of yield curve movements. Social implications The enhanced forecasting approach also equips the government with a valuable tool in setting macroeconomic policies. Originality/value 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.

Dynamic factor model for network traffic state forecast

2018 ◽  
Vol 118 ◽  
pp. 281-317 ◽  
Author(s):  
Tao Ma ◽  
Zhou Zhou ◽  
Constantinos Antoniou
Keyword(s):  
Network Traffic ◽  
Factor Model ◽  
Dynamic Factor ◽  
Dynamic Factor Model ◽  
Traffic State
Sign in / Sign up

Export Citation Format

Share Document