Using Induced Order Statistics to Construct Optimal Impact Portfolios with General Dependence and Marginals We develop a mathematical framework for constructing optimal impact portfolios and quantifying their financial performance by characterizing the returns of impact-ranked assets using induced order statistics and copulas. Our results apply to any joint distribution of impact factors and residual returns, making them broadly applicable to a wide range of contexts. We develop significant extensions of the theory of induced order statistics, with which we are able to characterize the distribution of residual returns of individual assets ranked by the impact factor. Our framework provides a toolkit for practitioners to construct impact portfolios and quantify their performance based on real data. This allows impact investors to achieve higher risk-adjusted returns than those with impact portfolios constructed using simpler heuristics such as negative or positive screening.