Bond Portfolio Management Using the Dynamic Nelson-Siegel Model

Factor models for the yield curve, such as the dynamic version of the Nelson-Siegel model proposed by Diebold and Li (2006), have been extensively applied to forecast bond yields. In this paper, we propose a novel utilization of this model in bond portfolio management. More specifically, we derive closed form expressions for the vector of expected bond returns and for their conditional covariance matrix based on a general class of dynamic heteroskedastic factor models, and use these estimates to obtain optimal mean-variance bond portfolios according to Markowitz's framework and to compute the value-at-risk (VaR) of portfolios composed of fixed-income securities. An empirical application involving a large data set of US Treasuries is presented.