http://www.bis.org/cbhub/list/author/author_6722/index.rss Research hub papers by author Clive G. Bowsher en http://www.dallasfed.org/research/papers/2008/wp0804.pdf Dallas Fed Working Papers by Clive G. Bowsher and Roland Meeks The Dynamics of Economic Functions: Modelling and Forecasting the Yield Curve 2008-12-01T12:00:00Z Abstract: The class of Functional Signal plus Noise (FSN) models is introduced that provides a new, general method for modelling and forecasting time series of economic functions. The underlying, continuous economic function (or "signal") is a natural cubic spline whose dynamic evolution is driven by a cointegrated vector autoregression for the ordinates (or "y-values") at the knots of the spline. The natural cubic spline provides flexible cross-sectional fit and results in a linear, state space model. This FSN model achieves dimension reduction, provides a coherent description of the observed yield curve and its dynamics as the cross-sectional dimension N becomes large, and can feasibly be estimated and used for forecasting when N is large. The integration and cointegration properties of the model are derived. The FSN models are then applied to forecasting 36-dimensional yield curves for US Treasury bonds at the one month ahead horizon. The method consistently outperforms the Diebold and Li (2006) and random walk forecasts on the basis of both mean square forecast error criteria and economically relevant loss functions derived from the realised profits of pairs trading algorithms. The analysis also highlights in a concrete setting the dangers of attempts to infer the relative economic value of model forecasts on the basis of their associated mean square forecast errors. The Dynamics of Economic Functions: Modelling and Forecasting the Yield Curve 2008-12-01T12:00:00Z Full text http://www.dallasfed.org/research/papers/2008/wp0804.pdf Clive G. Bowsher Roland Meeks Clive G. Bowsher and Roland Meeks 2008-03 Dallas Fed Working Papers C33 C51 C53 E47 G12 http://www.dallasfed.org/research/papers/2008/wp0811.pdf Dallas Fed Working Papers by Clive G. Bowsher and Roland Meeks Stationarity and the Term Structure of Interest Rates: A Characterisation of Stationary and Unit Root Yield Curves 2008-12-01T12:00:00Z Abstract: The nature of yield curve dynamics and the determinants of the integration order of yields are investigated using a benchmark economy in which the logarithmic expectations theory holds and the regularity condition of a limiting yield and limiting term premium is satisfied. By considering a zero-coupon yield curve with a complete term structure of maturities, a linear vector autoregressive process is constructed that provides an arbitrarily accurate moving average representation of the complete yield curve as its cross-sectional dimension (n) goes to infinity. We use this to prove the following novel results. First, any I(2) component vanishes owing to the almost sure (a.s.) convergence of the innovations to yields, vt(n), as n. Second, the yield curve is stationary if and only if nvt(n) converges a.s., or equivalently the innovations to log discount bond prices converge a.s.; otherwise yields are I(1). A necessary condition for either stationarity or the absence of arbitrage is that the limiting yield is constant over time. Since the time-varying component of term premia is small in various fixed-income markets, these results provide insight into the critical determinants of the stationarity properties of the term structure. Stationarity and the Term Structure of Interest Rates: A Characterisation of Stationary and Unit Root Yield Curves 2008-12-01T12:00:00Z Full text http://www.dallasfed.org/research/papers/2008/wp0811.pdf Clive G. Bowsher Roland Meeks Clive G. Bowsher and Roland Meeks 2008-08 Dallas Fed Working Papers