Abstract

We propose an a±ne term structure model which accommodates non-linearities in the drift andnvolatility function of the short-term interest rate. Such non-linearities are a consequence of discrete beta-distributed regime shifts constructed on multiple thresholds. We derive iterative closed-form formulanfor the whole yield curve dynamics that can be estimated using a linearized Kalman filter. Fitting the model on US data, we collect empirical evidence of its potential in estimating conditional volatilitynand correlation across yields.