Calibrating the Nelson-Siegel Class models and their Estimation using Hybrid-Genetic Algorithm Approach: A Case Study of Indonesian Government Bond

Muslim, Muslim and Rosadi, Dedi and Gunardi, Gunardi and Abdurakman, Abdurakman Calibrating the Nelson-Siegel Class models and their Estimation using Hybrid-Genetic Algorithm Approach: A Case Study of Indonesian Government Bond. European Journal of Scientific Research. ISSN 1450-216X / 1450-202X

[img]
Preview
Text
Calibrating the Nelson-Siegel Class models.pdf

Download (31kB) | Preview

Abstract

In this paper, we consider the problem of modelling the yield curve using Nelson- Siegel class models. Nelson-Siegel class models discussed here are NS, BL, NSS, RF and our proposed NSSE models. NSSE model is a model which extends of NSS by adding combination of first and second hump into third factor of this model, so that this extended model has four factors, namely the flat, the slope, the combination of first and second hump, and the second hump. The purpose of adding the additional combination of humps into model is to improve the accuracyof the Indonesia Government Bond yield curve model, which contain the high variation in the medium and long terms part of the curve. To estimate the models, we consider the hybrid-genetic algorithm (GA) and compare it withthe Sequential Quadratic Programming (SQP) method. We provide an empirical study using Indonesian Government-Bond Yield Curve (IGYC) data and we found NSSE model outperforms the other models. We further obtain the hybrid-GA approach performs better than SQP estimation method.

Type: Article
Subjects: L Education > L Education (General)
Depositing User: MUSLIM
Date Deposited: 06 Feb 2017 07:19
Last Modified: 06 Feb 2017 07:19
URI: http://repository.unja.ac.id/id/eprint/245

Actions (login required)

View Item View Item