Forecasting the Term Structure when Short‐Term Rates are Near Zero

Abstract

This paper compares the experience of forecasting the UK government bond yield curve before and after the dramatic lowering of short‐term interest rates from October 2008. Out‐of‐sample forecasts for 1, 6 and 12 months are generated from each of a dynamic Nelson–Siegel model, autoregressive models for both yields and the principal components extracted from those yields, a slope regression and a random walk model. At short forecasting horizons, there is little difference in the performance of the models both prior to and after 2008. However, for medium‐ to longer‐term horizons, the slope regression provided the best forecasts prior to 2008, while the recent experience of near‐zero short interest rates coincides with a period of forecasting superiority for the autoregressive and dynamic Nelson–Siegel models. Copyright © 2014 John Wiley & Sons, Ltd.

Forecasting term structure of the Japanese bond yields in the presence of a liquidity trap

“…Our findings are in contrast to some studies (such as Diebold & Li, 2006, and Nyholm & Vidova‐Koleva, 2012) that the NS‐based models outperform the RW when liquidity trap is not present. However, our results are largely consistent with respect to Steeley (2014), who reports that simple competitor models (in particular, the RW model) seem to provide better out‐of‐sample forecasts in the presence of liquidity trap.…”

“…Yu and Zivot (2011) find that the DNS factor with autoregressive lag order 1 model outperforms other competitors in the out‐of‐sample forecast accuracy, and the DNS factor state space model becomes appealing on the high‐yield bonds in the short‐term forecast horizons. Steeley (2014) engages both the DNS model and the less‐structured models to forecast the UK term structure when short‐term rates are near zero. It is found that the random walk (RW) and the AR(1) model have better forecasting performances.…”

The relative term structure and the Australian-US exchange rate

“…relative factors based on principal components, the random walk or AR (1) model of exchange rates) [11]. Such an extension would be along the lines of recent studies, notably Nyholm and Vidova–Koleva (2012) and Steeley (2014), which evaluate the out-of-sample forecast performance of contending models of the yield curve. .…”

“…2. Using the principal components approach of Litterman and Scheinkman (1991) to characterise the yield curve, it is possible to ascertain the medium-term maturity that gives the highest loading on the curvature factor (the third principal component among the yields) and then to infer for that maturity the value of λ that maximises the loading on the curvature factor in the Nelson-Siegel model. A good description of the principal components approach and how it relates to the Nelson-Siegel approach is provided by Steeley (2014).…”

Predictability of Emerging Market Credit Spreads Before and After Lehman Brothers: The Role of Macroeconomic Volatility

“…Since this benchmarking involves comparing non-nested models, we employ the Diebold and Mariano (1995) two-sided t-test which hypothesizes H 0 : The data set used for the riskless zero-coupon bond yield extraction and corresponding term-6 For further discussion, see Diebold and Li (2006) and Steeley (2014) In order to collect emerging market bond prices, we establish various eligibility criteria geared towards achieving reliable term-structure estimation. The rst requirement is the availability in each sample week of market price data on at least six Eurobond issues across a range of bond maturities (from 1 to 32 years).…”