Fat tails in leading indicators

Modeling the relation between the US real economy and the corporate bond‐yield spread in Bayesian VARs with non‐Gaussian innovations

“…For the BAA spread (which is mostly used as a leading indicator in this context, i.e., as a predictor for the real economy), we find that even after controlling for time variation in the second moment of the innovations, there is some scope for modeling non‐Gaussian behavior (cf. Kiss & Österholm, 2020). These findings are basically unchanged whether we consider univariate, bivariate, or trivariate specifications.…”

“…As pointed out already by Engle (1982), this could be caused by time‐varying volatility of innovations. However, there is also evidence that the innovations affecting the variables can benefit from being modeled with heavy tails (Brunnermeier et al, 2021; Chiu et al, 2017; Karlsson & Mazur, 2020; Kiss & Österholm, 2020; Liu, 2019).…”

“…Second, we make a general contribution to the literature on macro‐financial and macroeconomic modeling when it comes to the importance of time‐varying volatility and the presence of skewness and heavy tails. The issue of time‐varying volatility for macro‐financial linkages has been studied a fair amount in the literature—a closely related contribution is Alessandri and Mumtaz (2017), who use threshold Bayesian VAR models with stochastic volatility and financial regimes—but the same cannot be said about skewness and heavy tails, even if the latter has received an increasing amount of attention after the global financial crisis of 2008 (Acemoglu et al, 2017; Chiu et al, 2017; Delle Monache et al, 2021; Kiss & Österholm, 2020; Liu, 2019). Regarding heavy tails in the context of Bayesian VAR models, an important contribution is due to Chan (2020).…”

Modelling Returns in US Housing Prices—You’re the One for Me, Fat Tails

“…3 Fagiolo et al (2008) found the Laplace distribution useful when modelling the fat tails of GDP growth rates. The Student-t distribution has been used more widely in the empirical literature; see, for example, Cúrdia et al (2014); Clark and Ravazzolo (2015); Cross and Poon (2016); and Kiss and Österholm (2020).…”

Modelling Okun’s law: Does non-Gaussianity matter?

“…We consider two aspects of non-Gaussianity. The first of these is heavy tails (or "fat tails")-an issue that takes its starting point in the observation that many economic variables seem to experience large swings more frequently than what one would expect if the shocks hitting the economy are drawn from a Gaussian distribution; see, for example, Fagiolo et al (2008), Ascari et al (2015), Cross and Poon (2016), Liu (2019) and Kiss and Österholm (2020). The second aspect is that the unconditional distribution of many variables appears to be characterised by skewness.…”