Papers
Non-Linear Predictability in Stock and Bond Returns: When and Where is it Exploitable?
Co-authored with Masssimo Guidolin, David McMillan and Sadayuki Ono. Published in International Journal of Forecasting, 2009, 25, 373-399
We systematically examine the comparative predictive performance of a number of linear and non-linear models for stock and bond returns in the G7 countries. Besides Markov switching, threshold autoregressive (TAR), and smooth transition autoregressive (STAR) regime switching models, we also estimate univariate models in which conditional heteroskedasticity is captured by a GARCH and in which predicted volatilities appear in the conditional mean function. Although we fail to find a consistent winner/out performer across all countries and markets, it turns out that capturing non-linear effects may be key to improve forecasting. U.S. and U.K. asset return data are "special" in the sense that good predictive performance seems to require that non-linear dynamics be modeled, especially using a Markov switching framework. Although occasionally stock and bond return forecasts for other G7 countries also appear to benefit from non-linear modeling (especially of TAR and STAR type), data from France, Germany, and Italy imply that the best predictive model is often one of the simple benchmarks, such as the random walk and univariate auto-regressions. U.S. and U.K. markets also provide the only data for which we find statistically significant differences between forecasting models. Results appear to be remarkably stable over time, robust to changes in the loss function used in statistical evaluations as well as to the methodology employed to perform pairwise comparisons.
LikeCorrelation Dynamics between Asia-Pacific, EU and US Stock Returns
Co-authored with Don Bredin and Nghia Nguyen. Published in in S.-J. Kim and M. McKenzie (eds.) Asia-Pacific Financial Markets: Integration, Innovation and Challenges, International Finance Review, 2007, 8, 39-61.
This paper investigates the correlation dynamics in the equity markets of 13 Asia-Pacific countries, Europe and the US using the asymmetric dynamic conditional correlation GARCH model (AG-DCC-GARCH) introduced by Cappiello, Engle and Sheppard (2006). We find significant variation in correlation between markets through time. Stocks exhibit asymmetries in conditional correlations in addition to conditional volatility. Yet asymmetry is less apparent in less integrated markets. The Asian crisis acts as a structural break, with correlations increasing markedly between crisis countries during this period though the bear market in the early 2000s is a more significant event for correlations with developed markets. Our findings also provide further evidence consistent with increasing global market integration. The documented asymmetries and correlation dynamics have important implications for international portfolio diversification and asset allocation.
The Response of Industry Stock Returns to Market, Exchange Rate and Interest Rate Risks
Published in Managerial Finance, 2007, 33, 693-709.
This study investigates the sensitivity of stock returns at the industry level to market, exchange rate and interest rate shocks in the four major European economies: France, Germany, Italy and the UK. In addition to exposure to the market, significant levels of exposure to both exchange rate risk, in the four countries, and interest rate risk, in France and Germany, are identified. Further, responses to sources of risk are decomposed into components attributable to news about future dividends, real interest rates and excess returns. All three sources of risk contain significant information about future cash flows and excess returns.
Habit Formation, Surplus Consumption and Return Predictability: International Evidence
Co-authored with Tom Engsted and Stig Moller
On an international post World War II dataset, we use an iterated GMM procedure to estimate and test the Campbell-Cochrane (1999) habit formation model. In addition, we analyze the predictive power of the surplus consumption ratio for future asset returns. We find that, although there are important cross-country differences, for the majority of countries in our sample the model gets empirical support in a variety of diffrent dimensions, including reasonable estimates of risk-free rates, and the model dominates the time-separable power utility model in terms of pricing errors. Further, for the majority of countries the surplus consumption ratio captures time-variation in expected returns. Together with the price-dividend ratio, the surplus consumption ratio contains significant information about future stock returns, also during the 1990s. Finally, in most countries the surplus consumption ratio is also a powerful predictor of future bond returns.
Add Comment