Parametric estimation of different interest rate processes

Abstract

The paper examines the estimation of alternative interest rate processes describing the dynamics of UK interest rates. The methodology concentrates on selecting non-parametrically the number of autocovariances to use in calculating a heteroscedasticity and autocorrelation consistent covariance matrix. This is important for drawing correct statistical inferences. It is found that the dependence of volatility on the level of interest rates is not as high in the UK market as has been documented in earlier studies of the US market. Further results reveal that there was a structural change in the parameters of the interest rate process during the period of the participation of Britain in the Exchange Rate Mechanism (ERM) of the European Monetary System. However, by utilizing the proposed non-parametric schemes, it is shown that statistical inference is sensitive to the correct choice of the number of autocovariances.