ABSTRACT
THE LIBOR MARKET MODEL is a term structure model driven by multiplerandom factors. Owing to the multiple factors, the model has a rich capacity to accommodate rather general covariance structures for the forward rates. In other words, this model can be calibrated, or fitted, to various structures of covariance and various price inputs of benchmark derivatives, which is very desirable for a production model. Nonetheless, these advantages have long been hard to exploit. For a long time, the nonparametric calibration of the model to fit input correlations and input prices simultaneously had been an unresolved challenge to model users, and it had in fact become a bottleneck for the model’s applications. The situation was changed essentially in 2003, when a satisfying solution to calibration was introduced by Wu (2003). This chapter is devoted to the methodologies of Wu’s solution.
