Time series and cross-section information in affine term-structure models

In this article I provide an empirical analysis of the term structure of interest rates using the affine class of term-structure models introduced by Duffie and Kan. I estimate these models by combining time series and cross-section information in a theoretically consistent way. In the estimation I use a Kalman filter based on a discretization of the continuous-time factor process and allow for a general measurement-error structure. I provide evidence that a three-factor affine model with correlated factors is able to provide an adequate fit of the cross-section and the dynamics of the term structure. The three factors can be given the usual interpretation of level, steepness, and curvature.