Estimating a fuzzy term structure of interest rates using fuzzy regression techniques

Several papers in the financial literature propose using fuzzy numbers (FNs) to model interest rate uncertainty. However, in our opinion, the first problem to be solved is how to estimate these rates with FNs. In this paper, we attempt to provide a solution to this question with a method for adjusting the temporal structure of interest rates (TSIR) that is based on fuzzy regression techniques. This method will enable to quantify the anticipated rates in the fixed income markets for the future with FNs. In particular, we discuss how to estimate the TSIR with triangular fuzzy numbers because of their desirable properties. (C) 2003 Elsevier B.V. All rights reserved.