A distribution-free test of whether the dependence among asset returns is centrally (radially/reflection) symmetric, i.e. whether co-movements in the lower tail mirror those in the upper tail. The statistic is the Cramer-von Mises distance between the empirical copula and its survival counterpart; the test's asymptotic distribution is derived under stationary-time-series assumptions and critical values come from a tie-break bootstrap, letting it handle up to 25 series with about one year of daily returns. Applied to US portfolios it finds central asymmetry is time-varying, stronger in size-based portfolios, and concentrated in market downturns, consistent with financial contagion.
Proposes a nonparametric test for the null hypothesis that the copula of asset returns is centrally symmetric (equivalently radially or reflection symmetric). This matters because financial crises ra…