No abstract

We propose an Adaptive Dynamic Nelson-Siegel (ADNS) model to adaptively forecast the yield curve. The model has a simple yet flexible structure and can be safely applied to both stationary and nonstationary situations with different sources of change. For the 3-to 12-months ahead out-of-sample forecasts of the US yield curve from 1998:1 to 2010:9, the ADNS model dominates both the dynamic Nelson-Siegel (DNS) and random walk models, reducing the forecast error measurements by between 30 and 60 percent. The locally estimated coefficients and the identified stable subsamples over time align with policy changes and the timing of the recent financial crisis.

This paper addresses the issue of forecasting term structure. We provide a unifi ed state-space modeling framework that encompasses different existing discrete-time yield curve models. Within such a framework we analyze the impact of two modeling choices, namely the imposition of no-arbitrage restrictions and the size of the information set used to extract factors, on forecasting performance. Using US yield curve data, we fi nd that both no-arbitrage and large information sets help in forecasting but no model uniformly dominates the other. No-arbitrage models are more useful at shorter horizons for shorter maturities. Large information sets are more useful at longer horizons and longer maturities. We also fi nd evidence for a signifi cant feedback from yield curve models to macroeconomic variables that could be exploited for macroeconomic forecasting.traditional fi nance literature limits the information set to a number of observable yields and uses two alternative methods: extraction of latent factors via cross-sectional interpolation methods and extraction of latent factors by exploiting no-arbitrage restrictions.Among the cross-sectional interpolation methods, the Nelson and Siegel (1987) approach is the most popular. The Nelson and Siegel three-factor model explains most variances of yields at different maturities with a very good in-sample fi t. Diebold and Li (2006) have successfully considered the out-of-sample forecasting performance of this model by assuming that the three factors follow AR(1) processes.Among no-arbitrage models, the common approach is to assume a linear model for the latent factors and to restrict the factor loadings so as to rule out arbitrage strategies on bonds of different maturities. No-arbitrage restrictions serve not only for reducing the dimension of the parameter space, but also contribute to the theoretical consistency of the model. Dai and Singleton (2000) and Piazzesi (2003) have surveyed the specifi cation issues of affi ne term structure models in continuous time and discrete time, respectively. Duffee (2002) has shown the usefulness of essentially affi ne term structure models (A 0 (3) 1 ) in forecasting.These two approaches have been recently merged in an affi ne arbitrage-free Nelson-Siegel (AFNS) model (see Christensen et al., 2007;Le Grand, 2007), where the traditional Nelson and Siegel structure is modifi ed to rule out arbitrage opportunities.Models mentioned above are traditionally based only on the information contained in the term structure. Financial markets are clearly not insulated from the rest of the economy. Feedback from the state of the economy to the short-term interest rate is explicitly considered in the monetary policy reaction function introduced by Taylor (1993) and by now widely adopted to explain the behavior of central banks.Several papers indicate that macroeconomic variables have strong effects on future movements of the yield curve (among others, Ang and Piazzesi, 2003;Rudebusch and Wu, 2008). In particular, Ang and Piazzesi (2003) use an A 0 ...

This paper addresses the issue of forecasting term structure. We provide a unifi ed state-space modeling framework that encompasses different existing discrete-time yield curve models. Within such a framework we analyze the impact of two modeling choices, namely the imposition of no-arbitrage restrictions and the size of the information set used to extract factors, on forecasting performance. Using US yield curve data, we fi nd that both no-arbitrage and large information sets help in forecasting but no model uniformly dominates the other. No-arbitrage models are more useful at shorter horizons for shorter maturities. Large information sets are more useful at longer horizons and longer maturities. We also fi nd evidence for a signifi cant feedback from yield curve models to macroeconomic variables that could be exploited for macroeconomic forecasting.

We propose an Adaptive Dynamic Nelson-Siegel (ADNS) model to adaptively forecast the yield curve. The model has a simple yet flexible structure and can be safely applied to both stationary and nonstationary situations with different sources of change. For the 3-to 12-months ahead out-of-sample forecasts of the US yield curve from 1998:1 to 2010:9, the ADNS model dominates both the dynamic Nelson-Siegel (DNS) and random walk models, reducing the forecast error measurements by between 30 and 60 percent. The locally estimated coefficients and the identified stable subsamples over time align with policy changes and the timing of the recent financial crisis.