description- – This paper compares six term structure estimation methods in terms of actual ex ante price and yield prediction accuracy. Specifically, we examine the models' ability to price Treasuries for one to five trading days ahead. The models' performance differs markedly between in- and out-of-sample predictions. Their relative success also depends on time, the forecast horizon, and whether price or yield errors are compared. We examine the degree of loss in accuracy the modeler incurs by not using the best method: in particular, we compare the more complex splining methods and the parsimonious Nelson-Siegel model.
subjectcollectiondatepublishercreatorformat description- – This paper shows that the recent literature that tests for a long-run Fisher relationship using cointegration analysis is seriously flawed. Cointegration analysis assumes that the variables in question are I(1) or I(d) with the same d. Using monthly post-war U.S. data from 1959-1997, we show that this is not the case for nominal interest rates and inflation. While we cannot reject the hypothesis that nominal interest rates have a unit root, we find that inflation is a long-memory process. A direct test for the equality of the fractional differencing parameter for both series decisively rejects the hypothesis that the series share the same order of integration.
subjectcollectiondatepublishercreatorformat description- – Excess returns earned in fixed-income markets have been modeled using the ARCH-M model of Engle et al. and its variants. We investigate whether the empirical evidence obtained from an ARCH-M type model is sensitive to the definition of the holding period (ranging from 5 days to 90 days) or to the choice of data used to compute excess returns (coupon or zero-coupon bonds). There is robust support for the inclusion of a term spread in a model of excess returns, while the significance of the in-mean term depends on characteristics of the underlying data.
subjectcollectiondatepublishercreatorformat
