Optimizing the Cryptocurrency Investment Portfolio in Conditions of Uncertainty Using the Method of Data Envelopment Analysis - Robust Programming

Optimizing the investment portfolio is one of the vital issues in investment management. The various fluctuations of the financial markets and the uncertainty of the parameters make using classical models a severe challenge. Therefore, the optimization of financial models in conditions of uncertainty to adapt to the real world has been the focus of researchers. In the present research, a hybrid optimization model has been developed by applying data envelopment analysis and robust programming to assess risk with uncertain inputs and outputs. The statistical population of the research was extracted from the Coin Marketcap portal, where the updated data of the adjusted price of 37 selected top cryptocurrencies was used to estimate the risk and create an optimal portfolio. A two-step approach for selecting and optimizing the stock portfolio, increasing the stability of the investment process, and comprehensive evaluation of stocks based on financial criteria is proposed. In the first stage, the performance assessment of the selected stocks is done using the robust programming-data envelopment analysis (RDEA) method. Then, in the second phase, the investment amount in eligible stocks is determined using the half-variance average and absolute standard deviation models. The performance of the proposed approach is evaluated in a case study of cryptocurrency data with increasing uncertainty. The comparative results of robust peer models with two risk measures show that the mean semi-variance model performs better in choosing and optimizing the investment portfolio.