Volume 15, Issue 3

Forecasts generated by time series models traditionally place greater weight on more recent observations. This paper develops an alternative semi-parametric method for forecasting that does not rely on this convention and applies it to the problem of forecasting asset return volatility. In this approach, a forecast is a weighted average of historical volatility, with the greatest weight given to periods that exhibit similar market conditions to the time at which the forecast is being formed. Weighting is determined by comparing short-term trends in volatility across time (as a measure of market conditions) by means of a multivariate kernel scheme. It is found that the semi-parametric method produces forecasts that are significantly more accurate than a number of competing approaches at both short and long forecast horizons.

Testing for sustainability of public debt by analyzing how the primary surplus reacts to variations in debt, as suggested by Bohn (1998), has received great attention in economics literature. In this contribution, we apply that test to some countries of the euro area, including the PIIGS countries, where we allow for a time-varying reaction coefficient. We conclude that most economies are characterized by sustainable debt policies with the exception of Greece and possibly Italy. While Greece has clearly pursued an unsustainable debt policy, no clear-cut result can be obtained for Italy. For the latter country, the outcome crucially depends on the time period under consideration.

We propose a continuous-time process for modeling real exchange rates (RER) to provide new insights into the mechanism of reversion and into the limit properties of the process. In particular, in connection to the Purchasing Power Parity hypothesis, we discuss concepts such as mean-reversion, stationary distribution and expected time to leave certain intervals. Based on the proposed specification, we model eleven RER among industrialized countries and use the expected time to leave certain intervals to discuss further issues related to the stability of RER.

We present time series evidence on output convergence for 14 countries relative to the U.S. for the 1900-2008 period. We develop tests that allow for an unknown number of breaks in the series and also asymmetric convergence speed. We show that this asymmetry arises theoretically when the economy is not in the neighbourhood of the steady state. Breaks are modelled through a Fourier function fitted to the deterministic part of a time series, and asymmetries are modelled as a smooth transition function that matches the theory predictions. The tests are shown to have good finite sample properties. The results support the existence of convergence towards a mean that displays a break around the 1920s crisis and WWII. These breaks are associated to different patterns of post-WWII growth performance. Asymmetric mean reversion also appears to be an important feature driving convergence for most countries in the sample. This emphasizes the importance that disruptive historical events have on the dynamics of relative outputs.

The series on average hours worked in the manufacturing sector is a key leading indicator of the U.S. business cycle. The paper deals with robust estimation of the cyclical component for the seasonally adjusted time series. This is achieved by an unobserved components model featuring an irregular component that is represented by a Gaussian mixture with two components. The mixture aims at capturing the kurtosis which characterizes the data. After presenting a Gibbs sampling scheme, we illustrate that the Gaussian mixture model provides a satisfactory representation of the data, allowing for the robust estimation of the cyclical component of per capita hours worked. Another important piece of evidence is that the outlying observations are not scattered randomly throughout the sample, but have a distinctive seasonal pattern. Therefore, seasonal adjustment plays a role. We finally show that if a flexible seasonal model is adopted for the unadjusted series, the level of outlier contamination is drastically reduced.

The mixture structure of the generalized hyperbolic distribution of Barndorff-Nielsen (1997) is explored to quantify the contemporaneous correlation between return and volatility and to identify the effects of volatility feedback and risk premium within GARCH models. The statistical analysis of the excess returns based on the CRSP value-weighted portfolio index supports both volatility feedback and risk premium theories.