The small sample properties of tests of the expectations hypothesis: a Monte Carlo investigation

Abstract

In this paper, we extend results from the finance literature that explores small sample bias, due to persistent variables, in tests of present value asset pricing models. Using a Monte Carlo simulation approach, we investigate the finite sample behaviour of standard tests of the expectations hypothesis (EH) of the term structure, when interest rates are increasingly persistent. We document bias for the 'perfect foresight spread' regression and find that the variance bound statistic strongly tends to favour the EH when persistence is high. In a vector autoregression (VAR) framework, the Wald test and the slope of the ordinary least squares test using the 'VAR-theoretical spread' are strongly biased.

GDP volatility implication of tourism volatility in South Africa: A time-varying approach

“…International tourist arrival is preferred over international tourism receipts mainly because longer historical data are readily available for it. This is important because distortions due to small sample bias can be extreme for recursive estimators such as those used in this study (Garganas and Hall, 2011). Data on direct foreign investment inflows, trade openness and inflation are all sourced from the World Development Indicators database of the World Bank.…”

The Expectations Hypothesis of the Term Structure of LIBOR US Dollar Interest Rates

“…It is usually accounted for the existence of a time-varying term premium which is assumed constant in traditional tests. The other explanations include a small sample bias of the EH tests remaining severe in large samples, the over-reaction of long rates to current short rates as well as the asset pricing anomaly disappearing once it is widely recognized to the public (Tzavalis, Wickens, 1997;Bekaert et al, 1997;Garganas, Hall, 2011;Bulkley et al, 2011). It is also stressed that the predictive power of the yield spread depends upon monetary policies implemented by the central bank being much stronger at the times of monetary targeting than interest rates smoothing (see Mankiw, Miron, 1986;McCallum, 2005, among many others).…”