The study looked at how different copulas affect the accuracy of estimating Value at Risk (VaR) for a two-asset portfolio. They tested three types of copulas - Gaussian, Student’s t, and SJC - to see if the choice of copula matters. The results showed that while VaR estimates were similar, the tail dependence modeled by the Student’s t copula was not negligible. Comparing the Student’s t copula and SJC copula, they found that the symmetric tail dependence in the Student’s t copula was not natural. Overall, choosing an elliptical copula could increase the risk of errors in VaR estimates.