This paper studies asymmetric dynamics of real GDP growth by estimating linear and nonlinear quantile persistence over different parts of the conditional distribution for six major developed economies. Several novel quantile-based hypotheses are motivated in this paper and tested for the steepness asymmetry of real GDP growth that hypothesizes that contractions are steeper than expansions. The empirical results show that quantile persistence is generally high at far lower tails, thus requiring much longer half-lives to reverting negative deviations to the mean of real GDP growth and hence leading to gradual economic recoveries. By contrast, less persistence in far upper tails tends to generate sharp and short economic downturns that adjust positive deviations towards the mean of real GDP growth so as to cause abrupt economic recessions. In particular, this asymmetry in quantile persistence strongly supports the steepness asymmetry conjecture, robust to the presence of structural breaks and potential nonlinearities in real GDP growth.
机构:
Cornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USACornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USA
Zhang, Tao
Kato, Kengo
论文数: 0引用数: 0
h-index: 0
机构:
Cornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USACornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USA
Kato, Kengo
Ruppert, David
论文数: 0引用数: 0
h-index: 0
机构:
Cornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USA
Cornell Univ, Sch Operat Res & Informat Engn, Ithaca, NY USACornell Univ, Dept Stat & Data Sci, Ithaca, NY 14853 USA
机构:
Univ Toronto, Dept Stat Sci, Toronto, ON, CanadaUniv Toronto, Dept Stat Sci, Toronto, ON, Canada
Pesenti, Silvana M.
Millossovich, Pietro
论文数: 0引用数: 0
h-index: 0
机构:
City Univ London, Bayes Business Sch Formerly Cass, London, England
Univ Trieste, DEAMS, Trieste, ItalyUniv Toronto, Dept Stat Sci, Toronto, ON, Canada