This paper suggests an index-matrix approach to a knapsack-based portfolio selection model (E-IFKP) with parameters, characterized by elliptic intuitionistic fuzzy values. Elliptic Intuitionistic Fuzzy Sets are a tool to model the greater uncertainty of the environment, which is introduced in 2021. In the developed E-IFKP model, the price and the return value of the assets are determined by experts taking into account their rank. Three scenarios are proposed to the decision maker for the final choice -pessimistic, optimistic, and average. The proposed E-IFKP extends the dynamic programming approach for the Knapsack problem, which aims to select items to be placed in the knapsack to achieve the highest possible total value not exceeding its capacity. To determine the best option for an E-IFKP for certain data from the US stock exchange a software for conducting the proposed approach is developed and is used in the case study.