Copula methods for evaluating relative tail forecasting performance

Purpose The authors apply their method to analyze which portfolios are capable of providing superior performance those based on the Sharpe ratio (SR). Design/methodology/approach In this paper illustrate use conditional copulas for identifying differences in alternative portfolio strategies. SR. Findings results show that under Gaussian copula, both expected tail (ETR) and skewness-kurtosis exhibit remarkably low correlations respecting SR portfolio. This means these two different one. also find focus either upper (Gumbel) or lower (Clayton) render significant differences. short, copula analysis is useful understand what kind equity-screening strategy its corresponding measure (PM) performs better relation Practical implications Copula methods evaluating relative forecasting provide an tool when forecast very small found non statistically through standard tests. Originality/value Our evaluate models' because rather similar, conclusions statistical differences, can be defective as they may hinge subsample type size used, leading inefficient investment decisions. novel research topic.