Hi, I'm searching code to estimate real and nominal term structures in Ox (or Gauss). I like to construct a term structure model as proposed by Duffie and Kan. This class of models interest rates and prices of bonds are linear (affine) functions of a small number of factors. The dynamics of these factors are described by a generalized square root diffusion process. These models maken an explicit link between the time series dimension and the cross section dimension Estimation of the model is based on the available yields. Since the factors are treated as latent variables, they can be backed out using the Kalman filter. Estimation is then by maximum likelihood based on the conditional means and variances of the processes of the factors. To apply the Kalman filter in the estimation, the model has to be written in linear state -space form.
I really appreciate your help. Thanks in advance, Maurice. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
