Exponential-polynomial families and the term structure of interest rates (Q5933571)

Exponential-polynomial families like the Nelson-Siegel or Svensson family are widely used to estimate the current forward rate curve (some central banks use smoothing splines). These families depend on some finite-dimensional parameter \(z\). The author investigates exponential-polynomial families from an intertemporal point of view, when the plenty of cross-sectional data -- daily estimations of \(z\) -- are available. He considers the general exponential-polynomial model of Björk and Christensen and replaces \(z\) by an Itô process that is consistent with a given parametrized family of forward rate curves. These processes must have the property to provide an arbitrage-free interest rate model driving the parametrized family. They are characterized by the author in terms of their drift and diffusion coefficients with the help of the Heath, Morton and Jarrow drift condition. By solving an inverse problem the author obtains the characterization of all consistent Itô processes which is remarkably restrictive. Then the notion of consistency is generalized to e-consistency when \(\mathbb{P}\) is not a martingale measure. It is proved that e-consistent diffusion processes driving bounded exponential-polynomial families like Nelson-Siegel or Svensson ones are very limited: most of the factors are either constant or deterministic. As a conclusion, the author says that exponential-polynomial families are best not used for modelling the term structure of interest rates.