Empirical analysis and forecasting of yield curves
Date
2021-09
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this thesis, we focus on term structure models. An accurate estimate of the current term structure of interest rates plays an important role in many areas of finance. In addition, it is important to forecast the futures term structure. Therefore, a lot of research work is devoted to determining how to best estimate, model, and predict the interest rate structure. The first part of this thesis focuses on modeling and forecasting the yield curves. We used the Principal Component Analysis(PCA) , Nelson Siegel(NS) model, and Gaussian Regression Process(GPR) in order to fit and forecast the European yield curve with different maturities. The second part of this thesis focuses on the calibration of the term structure model, since calibration is a highly challenging task, in particular in multiple yield curve markets. We simulate rates for both single-curve and multi-curve frameworks using the exact method and Milstein method and then calibrate parameters of simulated rates using the Ordinary Least Square Estimation(OLSE) method and the Generalized Method of Moments(GMM).
Description
Keywords
Interest rate models, Principal Component Analysis, Generalized Method of Moments, Ordinary Least Square Estimation, CIR model
Citation
Dodampe Gamage, R. (2021). Empirical analysis and forecasting of yield curves (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.