Title: Evolving fuzzy modelling for yield curve forecasting

Authors: Leandro Maciel; Rosangela Ballini; Fernando Gomide

Addresses: School of Business and Accounting, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil ' Institute of Economics, University of Campinas, Campinas, Brazil ' School of Electrical and Computer Engineering, University of Campinas, Campinas, Brazil

Abstract: Forecasting the term structure of interest rates plays a crucial role in portfolio management, household finance decisions, business investment planning, and policy formulation. This paper aims to address yield curve forecasting and evolving fuzzy systems modelling using data from US and Brazilian fixed income markets. Evolving fuzzy models provide a high level of system adaptation and learn the system dynamic continuously, which is essential for uncertain environments as interest rate markets. Computational experiments show that the evolving fuzzy modelling approaches describe the interest rate behaviour accurately, outperforming traditional econometric techniques in terms of error measures and statistical tests. Moreover, evolving models provide promising results for short and long-term maturities and for both fixed income markets evaluated, highlighting its potential to forecast complex nonlinear dynamics in uncertain environments.

Keywords: evolving fuzzy systems; yield curve; rule-based models; interest rate; adaptive systems.

DOI: 10.1504/IJEBR.2018.091047

International Journal of Economics and Business Research, 2018 Vol.15 No.3, pp.290 - 311

Received: 25 Nov 2016
Accepted: 29 Mar 2017
Published online: 09 Apr 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article