Time-varying short-horizon predictability☆

Predictive regressions with time-varying coefficients

“…MSPE) and utility gains -across models for different periods related to the business cycle. Consistent with other recent papers (Henkel, Martin, and Nardari, 2011;Rapach, Strauss, and Zhou, 2010), we find significantly stronger evidence for predictability during recessions than during expansions (third row of Table 4) using both measures. The only exception, as it is not significantly different from zero, is the difference in MSPEs for models excluding time-varying coefficients.…”

“…Specifically, we expect the estimated risk premium to behave according to the dynamics implied by the Campbell and Cochrane (1999) model: it should increase during the recession and be larger at the end of the recession than at the end of the expansion. In such a framework, predictability would arise if our predictive models are able to anticipate the business cycle (see Henkel, Martin, and Nardari, 2011;Rapach, Strauss, and Zhou, 2010 for initial empirical support).…”

Stock market volatility and equity returns: Evidence from a two-state Markov-switching model with regressors

“…1 More recent research has employed regime-switching vector autoregressions to capture the richer dynamics of bond and stock returns and their predictors (Guidolin and Timmermann, 2007;Henkel et al, 2011) and Markov-switching multifractals to model regime switching at low, intermediate and high frequencies Fisher, 2004, 2007). Markov-switching multifractal models compare favorably with standard volatility models both in-and out-of-sample yet their attractiveness stems from parsimony as they are able to capture the distributional nonlinearities of financial data with a very limited number of parameters.…”

Technical market indicators: An overview

“…Prior literature has shown that some return predictability models' effectiveness varies across business cycles (Dangl and Halling, 2012;Henkel et al, 2011;Jacobsen et al, 2007) or across sentiment regimes (Stambaugh et al, 2012). Several of our market indicators exhibit sign-switching predictability across time, for example, the short interest ratio and weekly NYSE cumulative highs.…”

“…A recent strand in the literature documents that some return-predicting models are time varying and state dependent (Dangl and Halling, 2012;Henkel et al, 2011;Jacobsen et al, 2007;Yu and Yuan, 2011). For example, Jacobsen et al (2007) show that an increasing industrial metal index can significantly predict higher stock market returns during contractions but lower market returns during expansions.…”