Title: Do unequal-mass binary black hole systems have larger $χ_\text{eff}$? Probing correlations with copulas in gravitational-wave astronomy

Abstract: The formation history of binary black hole systems is imprinted on the distribution of their masses, spins, and eccentricity. While much has been learned studying these parameters in turn, recent studies have explored the joint distribution of binary black hole parameters in two or more dimensions. Most notably, Callister et al. [T. A. Callister, C.-J. Haster, K. K. Y. Ng, S. Vitale and W. M. Farr, Astrophys. J. Lett. 922, L5 (2021)] find that binary black hole mass ratio and effective inspiral spin $\chi_\text{eff}$ are anti-correlated. We point out a previously overlooked subtlety in such two-dimensional population studies: in order to conduct a controlled test for correlation, one ought to fix the two marginal distributions -- lest the purported correlation be driven by improved fit in just one dimension. We address this subtlety using a tool from applied statistics: the copula density function. We use Callister et al. as a case study to demonstrate the power of copulas in gravitational-wave astronomy while scrutinising their astrophysical inferences. Our findings, however, affirm their conclusion that binary black holes with unequal component masses exhibit larger $\chi_\text{eff}$ (98.7% credibility). We conclude by discussing potential astrophysical implications of these findings as well as prospects for future studies using copulas.