Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-17T17:08:14.564Z Has data issue: false hasContentIssue false

A yield-only model for the term structure of interest rates

Published online by Cambridge University Press:  26 November 2013

Şule Şahin*
Affiliation:
Department of Actuarial Sciences, Hacettepe University, Ankara, TURKEY
Andrew J.G. Cairns
Affiliation:
Department of Actuarial Mathematics and Statistics, Heriot-Watt University and Maxwell Institute, Edinburgh, UK
Torsten Kleinow
Affiliation:
Department of Actuarial Mathematics and Statistics, Heriot-Watt University and Maxwell Institute, Edinburgh, UK
A. David Wilkie
Affiliation:
Department of Actuarial Mathematics and Statistics, Heriot-Watt University and Maxwell Institute, Edinburgh, UK
*
*Correspondence to: Şule Şahin, Department of Actuarial Sciences, Hacettepe University, Ankara, TURKEY. E-mail: sule@hacettepe.edu.tr
Rights & Permissions [Opens in a new window]

Abstract

This paper develops a term structure model for the UK nominal, real and implied inflation spot zero-coupon rates simultaneously. We start with fitting a descriptive yield curve model proposed by Cairns (1998) to fill the missing values for certain given days at certain maturities in the yield curve data provided by the Bank of England. We compare four different fixed ‘exponential rate’ parameter sets and decide the set of parameters which fits the data best. With the chosen set of parameters we fit the Cairns model to the daily values of the term structures. By applying principal component analysis on the hybrid data (Bank of England data and fitted spot rates for the missing values) we find three principal components, which can be described as ‘level’, ‘slope’ and ‘curvature’, for each of these series. We explore the relation between these principal components to construct a ‘yield-only’ model for actuarial applications. Main contribution of this paper is that the models developed in the paper enable the practitioners to forecast three term structures simultaneously and it also provides the forecast for whole term structures rather than just short and long end of the yield curves.

Information

Type
Papers
Copyright
Copyright © Institute and Faculty of Actuaries 2013 
Figure 0

Table 1Descriptive Statistics for the Full Range of Days

Figure 1

Table 2Descriptive Statistics for the Full Range of Terms

Figure 2

Table 3Descriptive Statistics for Daily Differences for the Full Range of Days

Figure 3

Table 4Autocorrelation coefficients for Yields and Differences, for Full Range

Figure 4

Table 5Mean RMS and Maximum RMS for Different C Parameter Sets for Nominal, Real and Implied Inflation Spot Rates

Figure 5

Figure 1Loadings of the PCs for the Daily Hybrid Yield Curves

Figure 6

Table 6Lagged Correlations between the Monthly Yield Curves

Figure 7

Figure 2Auto-correlation Functions for the Nominal Spot Rates Residuals

Figure 8

Figure 3Auto-correlation Functions for the Real Spot Rates Residuals

Figure 9

Figure 4Auto-correlation Functions for the Implied Inflation Spot Rates Residuals

Figure 10

Table 7Residual Analysis of the Yield-Only Model

Figure 11

Figure 51-Month Ahead Forecasts with Upper and Lower Confidence Limits for Nominal Spot Rates (%)

Figure 12

Figure 6Difference Between the Observed and Forecasted Nominal Spot Rates (%)

Figure 13

Figure 71-Month Ahead Forecasts with Upper and Lower Confidence Limits for Real Spot Rates (%)

Figure 14

Figure 8Difference Between the Observed and Forecasted Real Spot Rates (%)

Figure 15

Figure 91-Month Ahead Forecasts with Upper and Lower Confidence Limits for Implied Inflation Spot Rates (%)

Figure 16

Figure 10Difference Between the Observed and Forecasted Implied Inflation Spot Rates (%)

Figure 17

Figure 11Fisher Relation Check for the 1-Month Ahead Nominal Spot Rate Forecasts (%)

Figure 18

Figure 12Errors for the Fisher Relation Check for the 1-Month Ahead Nominal Spot Rate Forecasts (%)